|
SuzukiSavage.com
/cgi-bin/YaBB.pl General Category >> The Cafe >> AMD & others --- Intel dominance in 2022 /cgi-bin/YaBB.pl?num=1597052560 Message started by Oldfeller on 08/10/20 at 02:42:40 |
|
Title: AMD & others --- Intel dominance in 2022 Post by Oldfeller on 08/10/20 at 02:42:40 OK, a new thread emphasis is needed as Intel is acting all "gone" at the moment so having Intel as a thread focus is sorta silly, right? Intel is still missing in action (for 4 weeks now) and AMD is still doing a weekly stream of BRAND NEW stuff that is both revolutionary and somewhat exciting. https://www.tomshardware.com/news/amd-patent-biglittle--hybrid-computing-implementation Scan this article as it is NEW STUFF you will see in future years. https://www.fool.com/investing/2020/08/08/why-amds-rally-could-be-far-from-over.aspx http://https://media.ycharts.com/charts/b8396848bfb65c54705cf6718c4871fb.png The gist behind this impressive AMD "stock value bump up rate" comes from AMD beating up strongly on NVIDIA instead of just beating up yet further on Intel. Every Ryzen 4000 APU processor that sells now means one less graphics card is needed, so NVIDIA gets down valued accordingly, as does Intel for not selling the CPU side of that Intel/NVIDIA matchup. AMD is going to hurt both companies with their Ryzen 4000 APU processors ..... not to mention what the 5nm Ryzen 5000 APU processor series will do to them in 2021. Intel is selling at $48 a share at the moment, about 45% of AMD's current price. =================================================== Right now AMD sees no need for a whole lot of Big-little in their processor mix, but AMD can see that need coming up in the 5nm future as the phone boys will roll up to become most of AMD's state of the art competition. Intel's recent foray into Foveros based Big-little was pretty much a complete flop so far, but that does not mean the phone boys will be nearly so clumsy about it --- they KNOW Big-little down deep in their bones after all and they also understand all the multi-layer lay down tricks you can easily do at TSMC.... For AMD to patent a chunk of this new tech specifically for PC processors way back in 2017 is sorta like a pre-emptive strike, sorta kinda like against technology thieves ..... and it is not like the Chinese phone boys don't do a lot of commercial espionage and jest plain tech stealing anyway so having a strong defensible patent on your PC scale Big-little from TSMC is jest smart business. |
|
Title: Re: AMD & others --- from AMD dominance in 2020 Post by Eegore on 08/10/20 at 12:09:33 "Every Ryzen 4000 APU processor that sells now means one less graphics card is needed," I've tried explaining this to people but I am not knowledgeable enough to explain the intricacies appropriately, or plain stupid questions for that matter. Is there a simple breakdown somewhere that you know of? |
|
Title: Re: AMD & others --- from AMD dominance in 202 Post by Oldfeller on 08/10/20 at 13:34:26 AMD Ryzen 4000 APUs include gaming level built in graphics layered right on the AMD APU's cpu unit with enough memory layered on to run the CPU and the GPU. You used to have to buy an Intel CPU and lots of system memory and a NVIDIA graphics card to get this same gaming functionality. The total count of Ryzen 4000 APUs sold will be approximately equal to the loss of the same count market share for both Intel and NVIDIA. ...... approximately equal, not exactly equal ....... but enough to hurt a lot. |
|
Title: Re: AMD & others --- from AMD dominance in 2020 Post by Eegore on 08/10/20 at 16:43:48 https://wccftech.com/intel-claims-gaming-real-world-benchmarks-leak-10th-gen-cpus-vs-amd-ryzen-4000-renoir-apus/ Obviously Intel has this handled right? My favorite part: "Intel is still relying on benchmarks to showcase the potential of its 10th Gen CPUs whereas its CEO stated that the industry should stop focusing on benchmarks." |
|
Title: Re: AMD & others --- from AMD dominance in 202 Post by Oldfeller on 08/10/20 at 22:04:37 Intel shows back up after a month of silence --- with Intel immediately getting caught out lying on their benchmarks again. The AMD Ryzen 7 4800H, on the other hand, is an 8 core and 16 thread APU on a 7nm process node which clocks in at 2.9 GHz base and 4.2 GHz boost. NOTE PLEASE, Intel's current best laptop chip (built at Samsung ???) is being compared to a last gen AMD APU instead of a current most modern gen AMD 4000 APU. Careful reading of the article also notes that the Intel unit is allowed by Intel specifications to pull over 80 watts of power during the actual test results collection -- Intel is SYSTEMICALLY CHEATING in other words, being significantly overclocked during the test collection period by the Benchmark Group people -- while the only slightly exceeding in performance of the AMD APU that consistently pulls less than 45 watts and is not being overclocked at all. I wonder how long a real Intel laptop can support an 80 watt thermal load (2x normal thermal loading) before the laptop is BIOS throttling all to hell and gone just to keep from smoking the CPU and the smaller separate GPU? 30-40 seconds mebbe ??? To run full duration of this benchmark at 80 watts would destroy the laptop, so it is fairly clear that this test is NOT being run off the cooling system of a common laptop. And who the heck is this Benchmark Group anyway, the ones who are cooperating with Intel on this very very tilted and tweeked testing? Intel always seems to like to use a little shill company for a front man when lying so blatantly on performance testing. However, all of this black bag manipulation tricks does is show that Intel is trying to defend their laptop market share from the onslaught of the latest generation of Ryzen 4000 APUs. Something must be hurting Intel ---- pretty badly I think. Will Intel be able to even try to lie effectively six months from now when faced with the first wave of 5nm Ryzen 5000 APUs at over 5ghz single thread performance at an even lower power draw? https://wccftech.com/amd-ryzen-5000-cezanne-mobility-apu-spotted-features-even-faster-vega-gpu-than-renoir/ =================================================== https://semiaccurate.com/2019/12/10/intels-benchmarking-antics-questioned/ RECOMMENDED READING This is not the first (or 10tth) time Intel has played black bag games with a "little no-name benchmark testing firm" . Read this for a timeline of all the Intel benchmark cheating leading up to the FTC ruling that what Intel has been doing with benchmarks was ILLEGAL ...... Intel then quit cheating for a little while, but in the last 6 months Intel has resumed all their old benchmark tilting and cheating efforts all over again. Both Apple and AMD have complained about this, but the FTC has not yet done any form of enforcement on their past decrees. Perhaps the fraudulent "Benchmark Group" shenanigans will prompt the FTC to go ahead and bust Intel for this latest direct violation. |
|
Title: Re: AMD & others --- from AMD dominance in 2020 Post by Oldfeller on 08/11/20 at 12:07:09 https://liliputing.com/2020/08/mozilla-lays-off-250-employees-a-quarter-of-its-workforce.html http://https://liliputing.com/wp-content/uploads/2020/08/firefox.jpg The corporation responsible for developing the Firefox web browser is laying off about a fourth of its workforce. As of 2018, the Mozilla Corporation had over 1,000 full-time employees. Now Mozilla has announced it’s restructuring the organization and laying off 250 people. Mozilla says the global COVID-19 pandemic has taken a toll on the corporation’s financial resources and downsizing the organization will help ensure it can continue operations. Among other things, that means closing down Mozilla’s operations in Taipei, Taiwan and in addition to the 250 jobs that are being eliminated, about 60 people will transition to new teams. Somewhat optimistically, the announcement also suggests Mozilla will try to do more with less by “acting more quickly and nimbly,” and being more open to experimentation, adjustments, and partnerships with outside organizations. |
|
Title: Re: AMD & others --- from AMD dominance in 202 Post by Oldfeller on 08/11/20 at 20:09:38 Let's talk about "layering" when building a CPU processor. Intel does it by clumsily stacking entire chipsets on top of each other with physical vertical solder connectors between them. 1-3 stacked chipset layers is all you can do with this trick and it is EXPENSIVE to do it at all. Defects quickly mount up when using Intel Foveros whole chip stacking techniques and the stacked chip heat transfer to the heat sink and up the fins and into the fan air flow are much much poorer. TSMC does it (at 5nm and below) by using two methods. First was the deep burn focused EUV (which has to be held down to 2-3 layers to keep the nagging build up of errors away) and second with the latest ASML machines that can put down fresh new layers of "over laid" metal coatings followed by top optical coatings and burning the chip all over again pausing on the places where they want a layer to layer connection to allow for burn through. This makes up a layered complex chipset that is not plagued with low yields like the previous methods. FIFTEEN 5nm functional coating/burn/coating vertically connected "metal layers" can now be done at TSMC with these newest sorts of tech. Sheets of fast L-1, L-2 or L-3 memory (or a heavy duty graphics processor or an AI co-processor or whatever else you need) can be added right on top of where they will be used on the main CPU itself. No sockets, no bus, no wasted space ...... and at the lowest cost possible and resulting in the fastest connection transfer speeds possible. This involves machinery that can actually lay down new burn coating layers while a chipset is actually being burned and built and burned again. This only works on deep burn focused EUV machines equipped with additional coating laydown functionality, machines that are only being built in by ASML (yes, some very expensive machinery to buy, but cheap to run once you own it and have a good deep burn design all worked out and debugged). AMD has designed an entire wafer load of chiplets using this 5nm process (actually they are on the 5th and final version of this chiplet development right now). Trial yields from the latest trial were over 80% which is a very good yield number for a brand new process. Trial run fully assembled AMD processors have been put together from these chiplets and then sent to the boardmakers to have all the assorted motherboard and BIOS stuff worked out completely. Once again, they are working on the final version now. http://https://static.techspot.com/articles-info/1840/images/2019-05-06-image-p.we This advance in lithography is integral to why Intel is losing out in an industry that they actually created in the decades past ..... and you will have to understand this layered deep burn lithography at least a little bit to even be able to follow the next few years of computer advancements. Read up here to get a primer on the subject. https://www.techspot.com/article/1840-how-cpus-are-designed-and-built-part-3/ |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/12/20 at 20:19:08 https://videocardz.com/newz/intel-tiger-lake-features-10nm-superfin-architecture https://liliputing.com/2020/08/intel-tiger-lake-laptops-are-coming-soon-if-this-hp-pavilion-leak-is-anything-to-go-by.html http://https://pbs.twimg.com/media/EfKTyA8U0AAji04?format=jpg&name=medium OK, we know Intel has just leaked 20 gigs of "top secret" stuff from the abandoned 10nm and 7nm processes that Bob Swan has just recently given up on, with Bob planning on buying a bunch of TSMC 6nm lithography built production instead. Some of this older style Intel homemade stuff will still come to market this year anyway (even if it performs significantly no better than some of the faster existing Intel 14nm PC chipsets) mainly because the 10nm process lines are already built now and the lines are up & running as best they can run (and in a bean picker's world such huge capital investments of that size must be run at full rate 24/7 in order to attempt to pay for themselves). Bob Swan has also said that all the folks he fired a few weeks ago "tended to overcomplicate things". Wow, what an understatement that was, Bob. I also understand more of why Bob Swan thinks that all these layers of these esoteric people simply lost track of "competitiveness" and spent all their time designing very very complex technically nice things that could not be brought into reality in a timely fashion on real world Intel equipment and this was why Intel has lost its way. I think Raja Koduri and Bob Swan holds the final responsibility for allowing this current EXTREMELY EXPENSIVE Tiger Lake "ongoing design error" cluster fook to go to build and Bob Swan needs to be let go as well right after he fires Raj Koduri for this sad non-competitive Tiger Lake mess that will launch on the second of next month. Intel Tiger Lake (available in units this fall) will swing only 1/2 the CPU core count of the recently released AMD Ryzen 4000 competition but Intel Tiger Lake will actually pull more power from the battery in order to run itself. The on board graphics of the two will be somewhat roughly comparable between the two brands (within a 40% advantage going to AMD) but with all the recent Intel cheating and black bag benchmark games that are going on right now in the industry trying to make a real comparison without using near identical laptops is nearly impossible to do. Really, until HP or Lenovo come out with the two chips running in the same exact laptop (battery screen and memory the same as close as can be done) a simple empirical test series on the two APUs will not be able to be done. :P ................ Recommendation: AVOID Intel 10nm and 7nm laptop chipsets like the plague as the sheer expense involved in making this 10nm + and ++ and +++ stuff will be astronomical compared to the already slightly superior and already much less costly AMD Ryzen 4000 processors |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/13/20 at 09:13:05 https://www.notebookcheck.net/The-Ryzen-7-4800U-powers-past-Tiger-Lake-U-in-Time-Spy-with-40-percent-higher-CPU-performance.457441.0.html Tipster @TUM_APISAK recently spotted a leaked 3DMark Time Spy benchmark listing for the Ryzen 7 4800U. The 15W 8-core, 16-thread Renoir SoC scored 1045 on the GPU test, but delivered a remarkable 6061 on the CPU test. This is nearly twice as high as what the Ryzen 7 3780U managed, and nearly 40 percent faster than leaked Time Spy CPU scores for an upcoming Tiger Lake U SoC. The leaked 4800U Time Spy scores cement the upcoming Renoir SoC's position as a potential game-changer in terms of power efficiency. Early reports indicated that the Ryzen 7 4800U would potentially offer double the battery life of competing Intel parts. It now appears that it'll also deliver significantly better performance to boot. The gains in both performance and efficiency largely stem from AMD's decision to use 7nm Zen 2 cores in its Ryzen 4000 mobile lineup. These offer significant IPC gains over the Zen+ cores in Picasso parts. However, the move to 7nm also allows for higher clocks and higher sustained performance. The Ryzen 7 4800U's 2x additional physical cores play a big role in better performance, too, with AMD managing to deliver 8 physical cores and 16 threads at a frugal 15W TDP. OK, compute power throughput and power draw to power the unit during the test can be easily measured and compared. Graphics throughput will require very same same constructed units doing the same tests in equivalent environments or by playing the same game over a same same endless loop. Same same units will be needed as I think the real world differences will be there, but in the end the results will be all muddled up by Intel's black bag benchmark tricks which are built into, well ........ everything ........ Cost to buy will be easy to compare, but also fairly easy for Intel to fudge in the very short term. Intel REALLY REALLY NEEDS a big widely reported win in the press here, boys and girls So, look to see some very serious pricing subsidy games being played and some very serious "environment tweeking" games and a little outright benchmark cheating going on (yes, all over again) to let old cheatin' Intel give out the temporary illusion of the big win that Intel so badly needs. =================================================== This Tiger Lake is the one must be seen a BIG WIN as the next Intel Lake that is coming, Alder Lake is a Big bigger from Intel that intentionally tries to go lower power and lower performance to get some much needed better battery life (Tiger Lake really sucks for battery life). This is Alder Lake stuff is NOT what you should be doing in a race where you are so far behind right now anyway. AMD has so many more physical cores that ARE so much more efficient that Intel is grasping at straws and making up specialty hardware just to show some sort of win --- any sort one sided win at the cost of something else. When AMD rolls out 5nm and 4 threads per core then AMD laptop products will process in massive parallel just like a mainframe does and the competition between mainframe processing and laptop processing simply isn't a competition any more on any front. Alder Lake is so retro and so inefficient it requires a BIGGER socket as it is a much BIGGER chipset. AMD sockets stay the same size (with much greater core counts) and get lighter and draw a lot less power. All these Lakes are simply Intel's old plans that are carrying forward in an environment where they have already failed. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/13/20 at 15:26:59 https://www.tomshardware.com/news/intel-alder-lake-x86-hybrid-cpu-for-desktops Intel is trying to make a case internally for Big-bigger for desktop chipsets. Just read it, it does not take long. When you read this, be thinking about TSMC's 6nm lithography two years from now and all the neat Big-little tricks the cell phone boys will be doing with ARM PC chip designs. I find it refreshing that AMD has taken out actual patents on Big-little for PCs that have johnny come lately Intel having to call theirs Big-bigger. It makes me smile some to see Chipzilla on their back foot to the phone boys for a change. =================================================== And this brings up my final question for the day ---- what is the compute industry going to do about Intel intentionally crippling Intel's competition using Intel's control of the industry BIOS library system (and the base system drivers, let's not forget them). We are not talking about Intel fixing things so it is working better for Intel's own products (it can't, it is stuck at 14nm and at a poorly working 10nm). Intel simply can't get any better than AMD. Intel has tried and tried and tried to get better --- and failed again and again and again. We are now talking about something different (and far worse). We are now talking about Intel intentionally screwing up the competitor's functionality REPEATEDLY week after week first with one processor library area followed next week by another BIOS settings area followed next week by a driver area. It must be sad that the only way you can try to win the race is to secretly trip up the other runners --- repeatedly no less. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/15/20 at 07:54:21 Spent some time off in rumor land this morning ..... there is interesting stuff coming up with 5nm AMD chiplets that are being run in trial lots as we speak. Threads per AMD physical core can be either 2 or 4 using 5nm, whichever makes sense with the OS and the intentional use it is to be put. OS products for mainframes can really use this sort of thread count but no common desktop OS can utilize it well right now. Linux can, though. IBM used to use up to 8 threads per physical core and they wrote that functionality into Linux, where it still remains. AMD 5nm cores are much smaller, so you can put about twice as many physical cores on the same AMD AM4 socket size if you downsize the center core from 14nm down to 7nm as AMD is starting to do. So, socket and die size stays the same, but functionality on an AMD AM4 socket chipset can just about double ...... Pin outs on the AM4 sockets are not being strained as so much of the I/O communications now takes place inside the AMD CPU assembly itself and only leaves it as a finished output stream. AMD chips with good on chip graphics don't use the bus for much except to handle inputs and outputs (no bus to graphics card or bus back & forth to memory takes place for much any more). 14 layer 5nm TSMC lithography means TONS of L1 and L2 system memory can be put down right over the CPU for fastest use. Raw compute speed is increasing with each new generation of chiplet. Ditto for AI co-processors and graphics layers. Close = fast and fast counts, you know. And this is all "right on top closeness" that is requiring NO BUS INTERFACE is making for the fastest memory and fastest graphics access yet. AMD has released some general info that a chiplet based CPU's cost is about half what a monolithic core CPU used to cost to do. Intel is behind the 8 ball as they are currently Foveros stacking 2-6 whole monolithic cores on top of each other, at a cost of like 4-10x as much as AMD's chiplet based processors. Yields on AMD's chiplet based processors are running at over 80% - yields of 90-95% is not uncommon on some AMD processors. And when you mess up a Foveros stack on an Intel processor you scrap out the 2-6 full sized chips that were used to build up the stack. Intel's yields are likely much lower with the Foveros stacking of full sized chips being somewhat error prone both in the stacking and in the complexity of what you are stacking. Rumor has it that Intel is done, that no matter what they come up with they can't do it fast enough to keep up with AMD and the phone boys who are moving 4-10x faster than Intel can move. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/15/20 at 16:43:34 I am not the only one who reads the tea leaves this way ...... https://ownsnap.com/amd-zen-4-ryzen-5000-desktop-cpus-might-use-l4-cache-4-threads-per-core-3d-stacked-architecture-according-to-strong-leaks/ http://https://ownsnap.com/wp-content/uploads/2020/05/zen-4-leaks-ownsnap.com-ryzen-5000.png Remember last year’s industrial leaks about AMD is up to something with it’s CPUs thread count? Exactly, SMT4 (Simultaneous Multi-Threading) feature may not come to Zen3 but it will be implemented for Zen 4 CPUs running on 5nm transistor shrink with matured strong architecture. It translates into 4 Threads Per Single Core feature that elevates overall performance up to 20-30% apart from the 40-60% bump the processors will be giving to processing power. Large IPC (Instructions Per Cycle) increase is also seen on horizon by 10-15% uplift in performance since memory bottlenecks can finally be reduced significantly using DDR5 fast access memory. Per SK Hynix claims, the DDR5 memory generation might run at 8400 Mhz effective speed which is a sweet dream of APU users. New AVX-512 instruction set is to be included for sure as previous leaks suggested. Novel chipset design optimized for Zen 4 futuristic architecture is inbound to Ryzen 5000 processors’ performance tweaking. More importantly, X3D chip packaging solution will be implemented too because AMD strongly presented 3D stacked method to solve many previous unsolved latency issues at it’s Financial Analyst Day two months ago. Obviously, 3D packaging allows billions of transistors getting together stacked in three dimensional positions to achieve maximum efficiency and performance. Larger Level 2 cache with a size of 1 MB instead of 512 KB per core is also supported by the latest leaks on Zen 4. Possibly, Level 4 cache appears to be featured in some Ryzen 5000 desktop CPU models but not all of the mobile processors. Current PC industry still utilizes Level 3 Caching method so having DDR5 super fast memory in future generation CPUs you might want to think that L4 cache is a necessary complement each other. Bigger cache sizes will be required to support 4 active threads per processor. Expect sizable Level 4 caches to be included on models specifically structured to support the heavy use of 4 active threads per processor for large data crunching tasks. Expect many layered processors at 5nm to support all of this L-1 through L-4 cache stuff. Also expect some AMD mobile APU laptops that can do jobs that used to require a Xenon Scalable workstation ...... ....... change, she comes ........ |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/17/20 at 03:54:00 Back from Rumor Land again ...... Time for a reality check --- Intel has only half the core count of AMD, addresses their systems with a PCIe 3.0 bus instead of a PCIe 4.0 bus like AMD uses --- Intel by and large totally lacks those very fast on-chip L-1-L3 operating caches. Lithography-wise Intel is at 14nm and AMD is at 7nm. Each company is working on their next generation of lithography but it is still clear that AMD is already two generations up on Intel and is likely to stay that way. AMD is less than 6 months away from their next lithography downsizing. Intel has not announced any defined reliable progress pathway to a new lithography level and Intel will have to buy chipsets from TSMC or Samsung whenever they are able to schedule enough free production time with TSMC or Samsung. Although Intel has some actual 10nm and 7nm equipment physically up and running, Intel has failed to make a good reliable transition to these new lithography equipment in a way that offers a great deal of advantage or value to the Intel customers. Intel is still running some small lots of their very latest processors at Samsung which are performance sorted, then acetone rag wiped and re-stenciled with appropriate Intel nomenclature for their various post sorted conditions ...... Intel is being very careful to not be telling anybody about the true state of their own home grown developments in 10nm and 7nm lithography ...... Intel is struggling right now to offer just the illusion of "being competitive" by widely advertising any small sliver of "win" that they can get. The things that Intel is willing to do to generate or create the illusion of a sliver of "win" is really somewhat shameful at this stage of things ........ Intel cheats benchmarks and plays games with gross overclocking and tweeked drivers and "harmful to others" BIOS libraries, etc. etc. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/17/20 at 12:41:28 https://www.fool.com/investing/2020/06/15/will-intel-new-chips-widen-moat-vs-softbank-arm.aspx Intel gestures towards the mobile market again with new chips for lightweight laptops and foldable devices. Smaller, faster, and more power-efficient Intel claims the Hybrid chips will fully support Windows 10 applications with a 56% smaller package area, 47% smaller board size, and a 91% reduction in standby power consumption than its comparable Y-series processors. In terms of real-world applications, Intel expects the chips to offer up to 24% better power efficiency during web browsing, up to 1.7 times better graphics performance, and up to 54% faster conversions of video formats. They could also offer double the throughput for AI-enhanced workloads. Intel is aiming these CPUs at lightweight laptops and foldable devices. Samsung's Galaxy Book S, Lenovo's ThinkPad X1 Fold, and Microsoft's dual-screen Surface Neo are the first three devices to use the new CPUs. Once again --- comparing Intel against Intel is both pointless and meaningless. Intel must do better than QUALCOMM in this market slice, not Intel. Intel must do better than AMD in this market slice, not Intel. Intel must do better than ARM Holdings X-series in this market slice, not Intel. Right now Intel has just now realized they have something to lose while they were deep in the throes of actually losing it ..... And I find it fairly funny that Intel is also "busy losing it" to Intel's business partner Mediatek, the same company Intel buys its radio tower modems from. And from Samsung, let's not forget good Intel business buddy Samsung, the one Intel has to build all of Intel's most modern chipset trial lots since Intel's own 10nm and 7nm is still acting fairly broken right now. Seriously, this market slice combat is really all about CHROMEBOOKS ...... with perhaps a side dash of Qualcomm Windows on ARM laptops. Chromebooks own the education market and are making inroads into business uses ....... Intel does not want to lose Chromebooks, and next year's light duty APUs from AMD may well steal that market segment away from Intel. ==================================================== Nice gesture, Intel. Now let's think about that astronomical price tag you are putting on your first Foveros stacked Big bigger samples that you are sending out right now ....... Intel, you may want to rethink that a little bit. Intel, you are defining the pricing of a market segment right now (intending for you to be the one to fill it) that is actually a "made to order" perfect fit for next year's lower cost higher performance Cezanne AMD 5nm APU's. Intel is having to compete on PERFORMANCE (Tiger Lake) Intel is having to compete on BATTERY LIFE (Alder Lake) With Intel, you can have one or the other ........ NOT BOTH AT THE SAME TIME With AMD 5nm APUs you get both, and you get a lower price tag and a lower unit cost and much better built in graphics. ::) When businesses make the wager on which company to use for processors when they plan a new product, they look for what is coming for the next 2 years. With Intel, you get a vague promise and a 50-50 chance of actually seeing it anywhere close to on time. AMD, 100% chance of seeing it on time, 25% chance of seeing it up to three months early. AMD contracts 2 years in advance for TSMC production slots, coming right in behind Apple in the order of build presidence for new lithography waves. AMD tends to hit their build slots on the nose with no issues building their products. ::) Intel wants to go with fully mature "maximum utilization" TSMC process waves, so Intel is coming in 2-3 years past the totally new wave front that Apple rides. AMD comes in between Apple and Intel time-wise and AMD costs less than either of them by far. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/19/20 at 16:41:18 http://https://static01.nyt.com/newsgraphics/2020/08/06/apple-mkt-cap/0cc4908d3731e64c440620d9276fefb2bad88477/apple-2-trillion-335.png Look at this and realize that people who used to blindly buy Intel are now blindly buying Apple. I suggested that Apple use part of this windfall to buy the rest of ARM Holdings (as a founder they already have a large chunk of it anyway) just to keep NVIDIA from messing it all up. Functionally, Apple already calls the development shots for ARM now anyway, so we would likely see no big change due to a more complete Apple ownership. Between contracting with TSMC and asking for the latest ASML technology to be brought forward and then buying the first of the new ASML lines and parking them at TSMC to do development and try outs and then also driving matching new ARM generations of Big-little processor designs to match up with the TSMC lithography generations ---- Apple already supports mobile technology development more than any other company does, both with their efforts and with hard cash to buy the hardware. Apple is also now driving the ARM based PC revolution in exactly the same way as Apple struggles to dump Intel completely. Apple does this for their own benefit, taking a exclusive use of the ASML machines that they buy until the next wave of new ASML machines that they buy are up and running. Apple has driven all of the progress in mobile that has taken place in the last 10 years following this roughly 2 year cycle. ;) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/22/20 at 03:17:14 https://liliputing.com/2020/08/lilbits-microsoft-brings-new-windows-subsystem-for-linux-to-older-windows-10-builds.html One of the most surprising things Microsoft did when the company introduced Windows 10 was to include a Windows Subsystem for Linux, which allows users to install Ubuntu or other GNU/Linux distributions and run them within Windows. The optional feature lets you open up a terminal window and run Linux commands without the need to reboot, switch computers, or launch a full virtual machine using software like VirtualBox. This year the Windows Subsystem for Linux got faster with the launch of WSL 2, which brings speedier file system performance, among other things. But not everyone is ready to update to the latest version of Windows 10 — so Microsoft has announced it’s making WSL 2 available for Windows 10 versions 1903 and 1909. That could come as good news for enterprise users or anyone else who may want to take things slowly when it comes to major operating system upgrades… but who would like speedier performance when running Linux commands within Windows. Ubuntu is starting to act like it considers itself a proper sub-set of Windows now-a-days. Microsoft flat relies on Linux in general for a lot of different forms of functionality now, between the Google browser functions and all the various pieces of Linux that MS has sucked up you can see where MS is going as they discontinue piece after piece of the old MS product and shut down and Linux replace these functions by updating them at night. MS intends to charge you a LOT of money over the years to repackage other people's stuff for you to use as "MS Windows 365". Hey, enjoy defragging your hard drive and fighting your nightly update installations and doing your bi-weekly anti-virus scans ....... :P REMEMBER, running plain pure Linux Mint Mate is faster by a lot. Carrying all the historical MS freight up on your shoulders is quite a load and it makes everything (including you) run slower ........ |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/25/20 at 12:40:00 https://www.amazon.com/PNY-CS900-120GB-Internal-Solid/dp/B0722XPTL6?ref_=Oct_DLandingS_D_44c1dcb8_60&smid=ATVPDKIKX0DER http://https://images-na.ssl-images-amazon.com/images/I/61KLO-goQiL._AC_SL1000_.jpg http://https://images-na.ssl-images-amazon.com/images/I/91FdfdADsiL._AC_SL1500_.jpg $16 is a hell of a good price for a solid state 120 gig SATA III hard drive. It is a limited time sale at Amazon, if you think you want one order it NOW (with a fall back plan to return it if you change your mind). https://www.amazon.com/PNY-CS900-120GB-Internal-Solid/dp/B0722XPTL6?ref_=Oct_DLandingS_D_44c1dcb8_60&smid=ATVPDKIKX0DER This drive is SATA III natural but is 100% back compatible to SATA II (automatically no less) which means you can plug it in as a boot drive on just about any SATA equipped computer that has SATA cabled drives in it already. This drive works in stuff going back 10-15 years or more, but you can check your old machine specs on the internet fairly easily if you feel doubtful. Look for your drive specs and look for SATA II or SATA III. YOU WILL FEEL THE SPEED if you are coming off a spinning platter hard drive. I have personally used one of these (mine cost a bit more a year ago) for almost a full year now and I have nothing but complements to say for the cheapest best easiest performance upgrade to an older PC that I have ever done. YOU CANNOT BUY A SPINNING PLATTER DRIVE THIS CHEAP period, so maybe it is time to go get a bit more modern with your boot (OS running) drive. You can keep your old platter drive as Drive D: or E: or whatever, keeping all the access to programs and data that is on the old drive. It acts just like a SATA drive, plugs in with the same cable and everything. It was intended as a laptop drive, so guess what, it fits laptop mounting points. In a big desktop unit, I simply velcro'd mine on top of my old drive bay. So thin tiny, small and light is easy to put in the big case machines with velcro, screws or whatever. Being solid state, orientation is meaningless and duct tape would be enough to hold it up (if you had to fall back on duct tape just because you love the stuff). |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/27/20 at 04:07:41 https://www.youtube.com/watch?v=qTzUAdUx3T4 it is a YouTube, click on it Today, we take a look at what you can do with a Raspberry Pi 4 using the TWISTER alternate operating systems (yes, there is more than one Apple based OS clone look-a-like that is able to be switched between MAC OS types at will that is built into this particular Raspberry Pi 4 OS). This alternate does MAC for those who think they would like to try a MAC OS clone. What struck me was how complete the clone was and how FAST it ran on the Raspberry Pi 4. We get all get stuck on trying to use Windows for comparisons ---- and Windows is always notably slow on a Pi. MAC is much faster anyway, so the Pi cone for MAC is much much faster --- it looks pretty useable actually. WHY IS THIS WORTH WATCHING? You need to expand your mind as far as MAC and Raspberry Pi 4 goes ...... this TWISTER stuff is approaching the realm of full Mac on PC functionality. Everything you see here is actually Linux showing the facade of something else -- Linux and Mac are very close to each other anyway so this level of speed and functionality should not be a surprise to anybody. What surprised me was the response speed and performance coming off of just 4 each A-72 style Amlogic cores ..... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/27/20 at 17:36:46 As you read Intel press releases about Tiger Lake, remember Tiger Lake is inefficient, hot running and very power hungry compared to AMD's new laptop chipsets. Also remember Tiger Lake has half the physical cores of a current competing AMD APU chipset. Intel will want to compare its newest Tiger Lake against an AMD APU from this past spring if it compares itself to the competition at all. Also remember any Intel benchmark comparisons are based upon Intel boosting its processor to 80 watts during overclocking runs using very non-laptop cooling methods. What is needed is independent tests using same same screen size, chassis and battery in a real laptop to do a model to model comparison if you want to talk real graphics power comparisons ...... and any laptop performance characteristics must run the same exact programs off a charged unit battery in a loop length that is long enough to end any Intel "magic minute" tricks. ::) In short, Intel can only win by cheating and this has been true for 6 months now ...... =================================================== Watching the tame Intel reviewers on YouTube going over Tiger Lake is like an exercise in propaganda management. How they dance and weave to avoid the negatives is almost like artistry -- never do they mention that AMD has twice as many processor cores as Tiger Lake and that Tiger Lake sucks a third more power out of the battery to simply run at equivalent graphics levels to AMD, and note that Tiger Lake is never compared to a current AMD Renoire APU chipset, although that is what it is selling against. A favorite trick is to compare Intel to an AMD unit that has the same core count (if that is possible to find that is). In lots of cases the reviewers have to go back in time to find an older AMD unit to get that core count equivalency. And who wrote the rules for this "core count equivalency" BS anyway? Do you remove cores from your processor when running programs? AMD uses its higher physical core counts and doubles that physical core count with 2x threads per core in order to offer the end user more speed and more power for roughly the same money (or less). And the whole "core count equivalency" exercise is completely asinine as AMD has really got twice the physical cores of Intel comparing the two companies up and down the current line ups. AMD Has better graphics. And AMD COSTS LESS to buy. Intel's Youtube shills continue to work hard to con their Intel fanboys into paying half again more money for inferior Intel performance --- Intel's Youtube shills work even harder to con them even more to get them to feel good about having "Intel Inside". |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/29/20 at 07:51:04 https://www.youtube.com/watch?v=ZYqG31V4qtA Here is an old Intel fanboy named Linus who has converted to being an AMD fan ---- why? Because he simply had to. Inescapable hard facts drove him to it. ;D Here is another somewhat older fact based analysis: https://www.youtube.com/watch?v=4V3uB12mRrU Take one of these as a temporary "antidote pill" for when you get blindsided by some blatantly biased Intel pre-release propaganda. Then, you will need to wait a day or five until the non-intel people can weigh in on the specific dose of Intel BS you were just fed by Intel Marketing. REMEMBER, nobody has seen the actual hardware yet --- just some Intel patronage "testing reports" from little bitty paid shill companies (or from some really stupid YouTube people who are just reading Intel spec sheets and reporting it as news, some of which aren't even getting paid by Intel ---- they are just so stupid they actually believe Intel's propaganda as written). |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 08/31/20 at 21:05:24 http://https://i.pcmag.com/imagery/articles/016yXQZCZtDGZxmUfpB8pav-3.fit_scale.size_850x478.v1578447684.jpg Whoever built the new chipsets and re-packaged them on the smaller daughterboard did a "better than Intel" job of it ....... (Samsung ????) Looks like an improvement in packaging and construction, but soon we will have real production bult units out in the wild for real independent testing to begin. Intel has lost all credibility lately and has people lined up to debunk all their BS claims about this one just like all the other Intel innovations of the last year ----- innovations that were not what was promised at all. =================================================== Tiger Lake has landed. Tiger Lake is the best Intel laptop chipset ever. Tiger Lake can actually try to compete with AMD's latest and greatest ---- but can it win in any clear fashion ???? Folks doing testing are seeing a very close match up using the existing "real world" functional tests as proposed by Intel but only one thing still rings true ---- due to more cores doing more work AMD is being very hard to beat on any form of data crunching in these tests. AMD is more energy efficient and gives better battery life. Intel runs hotter and drains a laptop battery faster. Intel remains hard to beat on laptop gaming where Intel owns all the metrics libraries and BIOS drivers that are all used by popular games (but this is really murky ground as you can't really see past the metrics/BIOS software aspects of things). Pro-Intel pundits are saying you can't really use existing benchmarks as the new Intel compute structure apparently moves more data per clock cycle and most all of the existing benchmarks assume a set of comparatively steady old throughput per cycle numbers and that is an invalid assumption now a days. Both AMD and Intel have increased instructions per clock cycle structurally (intel is using AI functions calls do this, AMD has better lithography and more cache memory) AND both have sped up their clock speeds accordingly as well. This "functional compute swell" is worth more than the current lithography downsizes (for right now, anyway). Now you can see even Intel is now saying some sort of unbiased group effort is needed by inference to make up a new set of performance benchmarks ....... =================================================== COMPETITION between Apple vs Intel vs AMD has caused the world to MOVE appreciably in the good direction. This last Intel progress will supposedly fire across 150 different laptops over a dozen brands by Christmas time of this year. Apple will release new very expensive competitive products at this new level. AMD already has equivalent products that are already released months ago, with new structure and new lithography levels outcoming every six months or so. AMD is making progress at the same rate as the phone boys do --- or roughly twice as fast as Intel does. Apple and AMD will step down to 5nm shortly, new turf where Intel currently cannot go. Because of Samsung and TSMC are acting as foundries for Intel, expect Intel to be at 6nm within a year and a half going with laptop chipsets first. Intel has gotten off their dead asses and moved ..... finally. Now expect Intel to move to 6nm with their entire product line, using laptop chipsets as their entry wedge. An eight core (two threads per core) is now the new minimum in computing. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller on 09/03/20 at 06:07:50 https://liliputing.com/2020/09/qualcomm-snapdragon-8cx-gen-2-5g-is-designed-for-pcs-with-25-hours-of-battery-life.html http://https://liliputing.com/wp-content/uploads/2020/09/8cx-gen-2_05-700x394.jpg According to Qualcomm, its new chip can power fanless PCs with more than 25 hours of battery life, while offering “up to 50% greater total system performance than the competition” according to the company’s own internal testing. http://https://liliputing.com/wp-content/uploads/2020/09/8cx-gen-2_02-700x393.jpg Qualcomm says compared to an unspecified 10th-gen Intel Core i5 processor running at 15 watts, its 7 watt Snapdragon 8cx Gen 2 5G chip offers up to 18-percent better performance and up to 39-percent better efficiency. Things look even better for Qualcomm when the company pits its chip against a 7 watt Intel Core i5 chip with Hybrid Cores (most likely a Lakefield processor). In that case, Qualcomm says the Snapdragon 8cx Gen 2 5G is up to 51-percent faster and up to 58-percent more efficient. Qualcomm has also updated the AI engine from Hexagon 685 to Hexagon 690, which could bring a small performance boost in some tasks. Both the Snapdragon 8cx 5G and the Snapdragon 8cx Gen 2 5G feature: 8-core Qualcomm Kryo 495 processors 8-channel LPDDR4x RAM Support for NVMe and UFS 3.0 storage Qualcomm Adreno 680 Extreme graphics Qualcomm Qualcomm Spectra 390 image signal processors (support for 32MP single cameras or 16MP dual cameras) Qualcomm Snapdragon X55 5G modems (mmWave and sub-6GHz support) Qualcomm Snapdragon X24 LTE modems Qualcomm Quick Charge 4+ support The company says the first devices with a Qualcomm Snapdragon 8cx Gen 2 5G chip should ship in late 2020 and if you’re wondering what they’ll look like, Acer’s got you covered. The company’s upcoming Acer Spin 7 (SP71) is the first PC announced to feature the new chip. Qualcomm is not really seen as actually leading any industry wide groups lately, but they do point out that when you want to talk quiet & light (fanless) laptops (the ones that last well over a whole day between charges) Qualcomm is the current technology leader. Apple and Qualcomm are both using modified versions of ARM Cortex N-1 cores so ARM is in the technological center of the low power high performance side of things. AMD is using some ARM technology as well, but AMD has reworked theirs significantly over towards the x86 side of things. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/04/20 at 16:59:21 https://www.pcmag.com/news/intels-tiger-lake-laptop-cpus-slightly-faster-much-smarter ...... this grinning goon Intel guy apparently thinks it is all pretty funny, to him anyway. http://https://i.pcmag.com/imagery/articles/024g5C7n84y6VoAJJYDfH8d-6.fit_scale.size_850x478.v1599093830.jpg Some of the smoke and mirrors is beginning to clear finally ---- Tiger Lake isn't really any faster than AMD Renoir or Apple or Qualcomm, Tiger Lake just carefully uses a lot more of the newest AI function calls that have been quietly built into the systems libraries behind the test software that is being run in these comparisons. These are new AI calls that really only play well for Intel right now (and notably not at all for AMD or APPLE) until AMD and Apple decide to actually implement these new AI function calls in their base softwares. Harsh folks will say Intel has simply gotten caught at widespread background cheating BIG TIME yet again ....... me, I think that is too simple an explanation. Intel cannot move their lithography base like the others have done, so Intel is attempting to change the entire conversation completely into something that Intel can actually do. Intel releases new versions of its Core laptop processors every year, and most years the company tends to talk up how much faster they are than the previous generation. But Wednesday's launch of its newest silicon, the 11th generation “Tiger Lake” chips, is different. Intel’s main claim with Tiger Lake is not that it’s faster than the 10th generation “Ice Lake” family (though it is), but that it’s faster than the latest Ryzen laptop processors from AMD. Why Intel's shift, deigning to mention competing silicon from its archrival? The obvious answer is that the latest Ryzen 4000-series “Renoir” chips have been trouncing Ice Lake in raw performance on classic benchmarks ever since the first Renoir laptops went on sale earlier this year. So regaining the lead is important, though we’ll have to take Intel’s word for it since no Tiger Lake laptops are available for independent tests just yet. But the more interesting answer is that the persistent delays in the research, development, and manufacturing of Intel silicon that helped AMD get a leg up in the first place have led Intel to explore other ways of making their processors faster that don’t rely on pure silicon-muscle improvements. Perhaps the most significant of these methods is adding artificial intelligence (AI) capabilities, which are responsible for much of what backs up Intel’s claim that Tiger Lake is once again the best-performing processor family for thin-and-light laptops. Speeding Up Those Crops and Blurs Enabling AI capabilities on a PC or Mac is a team effort involving the chip maker, the computer manufacturer, and the developer of the app that stands to benefit from AI. In Tiger Lake’s case, Intel has added support and optimizations for numerous AI platforms that manufacturers and developers can use to speed up the performance of their apps. http://https://i.pcmag.com/imagery/articles/024g5C7n84y6VoAJJYDfH8d-2.fit_lim.size_768x99999.png As Intel dumps bunches and scads of AI speed up calls into the various benchmarks tests and into the BIOS library functions that they know the competitive tests will use --- so look to see AMD and the others simply start intentionally using these AI speed up calls as part of their own arsenal. Intel does not have a lock on AI functions, arguably Apple and AMD are just as far along in implementing AI in hardware as Intel is, actually. Intel has simply made an early user play as the first mover and has identified a new way to make significant benchmark "progress" in a way that is somewhat sneaky, but is arguably worth pursuing by everybody. And these AI functions and calls are very real right now and should be used more by all softwares anyway ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/08/20 at 00:32:20 https://www.youtube.com/watch?v=aFHBgb9SY1Y Ok ....... Reviewers are beginning to post debunks of Intel's Tiger Lake smoke and mirrors. You can watch while this while the guy goes over various Intel BS smoke items and the inverse reflections of the same things that are also being claimed by Intel as "yet another Intel victory". He stumbles a lot over what he is trying to say because it is all so confusing and obtuse, but that is what is going on in reality with Tiger Lake's BIG Progress. Point was made by the dude that Intel was repeatedly bashing AMD as being "an imitator" when actually the opposite is true. Intel is well behind AMD and Intel is the one imitating AMD's advancements and is actually cheating to do it. :o ...... once AMD starts using all these cute AI calls in their own test runs, only then will AMD be acting as an Intel imitator. Once you back out Intel's much greater use of AI background functions (support for these calls were just recently installed in the benchmark's support libraries by Intel) then all Intel has in Tiger Lake apart from AI is a WHOLE LOT of hot air and PR and advertising. http://https://i.pcmag.com/imagery/articles/024g5C7n84y6VoAJJYDfH8d-2.fit_lim.size_768x99999.png Intel makes a lot of noise about "not using benchmarks" ---- instead they use AI calls to their recently installed background library resources to greatly speed up the execution of SOME of the old benchmark function items that they say they are not using ...... somewhat confusing, right ??? ::) Point being proven is that the AI being actually used is very very fast (but the phone boys have known this for years and them phone boys can cheat much better than Intel can, trust me on that one). I also dis-liked the fact that Intel has just jumbled up all their existing chip naming nomenclatures on all their old chipsets so they can now present an old existing 14nm product as being a brand new product that is "a whole lot faster now" because the cheating AI metrics calls make it appear that way. In short, AMD is doing all their stuff by the rule book and actually is using increased hardware speed to make the real progress that they show, while Intel has blown so much AI smoke up everyone's arse that by now all the reviewers are simply choking on the smoke as they burp it up ....... Yes, AI is neat, but Intel's saying their use of new AI calls "is your Tiger Lake processor being faster" is simply false logic --- well, actually that's just Intel lying out their arse a bit, lying yet again. :P |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/09/20 at 04:37:11 https://www.pcworld.com/article/3573774/intel-piles-on-the-benchmarks-to-show-tiger-lake-is-the-fastest-laptop-cpu.html Mark Hachman, Senior Editor of PCWorld magazine has written a short piece reminding everybody that NOTHING IS KNOWN until Intel actually ships real units that can be independently tested. http://https://images.idgesg.net/images/article/2020/09/intel-graphics-gameplay-100856522-large.jpg Whenever Intel, AMD, or Qualcomm launch a chip, the first question anyone asks is, how fast is it? We won’t know how Intel's Tiger Lake performs until we test it ourselves. But Intel certainly believes its 11th-gen Core chips are much faster than AMD’s mobile Ryzen 4000 parts—because Intel has made claim after claim to that effect. Intel talked aggressively about Tiger Lake’s performance early on, but the comparisons that accompanied the formal launch of the Tiger Lake family Wednesday were, well, sort of wishy-washy. It wasn’t until later in the day, during deeper-dive press briefings, that Intel began hammering home Tiger Lake’s performance compared to AMD’s mobile Ryzen. Let’s be clear: These are Intel’s claims, using its own microprocessors, laptops, and benchmarks. Are they objective? No. We’ve already seen crazy-cheap notebooks like the mobile AMD Ryzen-powered HP Envy x360 13 perform exceptionally well for their price and weight. Intel has a big stake in reclaiming its dominance in notebook PCs—even though AMD’s mobile Ryzen 4000 is now about nine months old. Mark goes through the long list of "Intel victory claims" noting in passing just how useful (or not) the AI performance boost being talked about may be to real laptop users. It is clear that AI is claimed to be useful in carefully tuned situations but we all have to agree with Mark that Intel's intentionally confusing carefully tuned "certain item only" AI boosts with their new processor's actual general overall real performance is simply an attempt to sell yellow snow to skeptical Eskimos. And all the skeptical Eskimos are ALL PROMISING that real testing will take place with purchased units just as soon as any become available. And pray tell, why again are the independently produced Tiger Lake units not here yet? Having some problems with Intel's first real production runs, eh ........ speak louder --- say that again Intel boys? ::) We are all very curious to see a real laptop that can cool and support an 80 watt loading during these benchmarking runs on a mobile chipset laptop. ::) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/09/20 at 06:11:45 https://www.pcworld.com/article/3573639/why-intels-tiger-lake-cpus-will-make-laptops-more-confusing-to-buy.html http://https://images.idgesg.net/images/article/2020/09/intel-greg-bryant-tiger-lake-100856440-large.jpg How it can be that two Intel 11th-gen Tiger Lake notebook PCs, with exactly the same chip inside, can offer performance that differs by up to 37 percent? And more importantly, how can you, the buyer, tell the difference? The short answer is: You can’t. An unfortunate side effect of how Intel develops processors like Tiger Lake will make it more difficult for consumers to determine the actual performance of a laptop just by reading a list of its features, and it will probably get harder over time. The problem is that Intel is designing mobile microprocessors like Tiger Lake to allow laptop makers greater flexibility in how they choose to “clock” them, or at what frequency they’re assigned to run. A laptop that uses a specific processor could run 37 percent faster than a second laptop that includes exactly the same chip, Intel executives say. That means laptop buyers will have to look harder at actual performance reviews with comparative data to understand what they're getting. The problem actually cropped up earlier, as Intel moved into its prior (10th-gen) Ice Lake chips. But Intel executives highlighted it during its Tiger Lake launch as something to watch out for. How, you ask? By downplaying it and then obliquely lying about it ..... ::) So ....... Intel is saying that the laptop housing's total heat removing capacity determines everything about its builder's relative performance choices compared to AMD based units that are simply much cooler running ???? That Intel's HEAT shutdown limits are the builder's choices when he sets the Intel processor's maximum output level for all time. Once picked by the builder, the shutdown limits remain forever. Next, during the same run of the same model laptops the different machines can get hot differently, so they each can have different limits set for them individually. :P ...... so, are they saying that some Tiger Lake machines are lemons and will always suck? So, Tiger Lake runs just that hot in normal day to day use, huh? Folks are now thinking that a Tiger Lake Intel based laptop machine's "watt rating" has more to do with the laptop housings ability to dump heat based upon this new understanding of "watts as heat dumped" ----- and not any "normally rated" Tiger Lake power usage can be expected to play across the board. And we are also sensing a great deal of in processor performance variation is going on with Intel's 10nm Tiger Lake product. Both a huge potential run-to-run variation and a quite large ~40%~ level variation part to part within the same wafer. Intel has also used AI selectively to speed up certain functions used in the common benchmark tests, then has reported this as "increases in processor throughput". It is not, it is hidden use of AI that was in essence staged since Intel controls many of the background x86 libraries that all softwares utilize. It is interesting that these speed ups only work for certain Intel machines since Intel does not want competitors to look good, just them. WHY HAS INTEL PAINTED THEMSELVES INTO THIS NETWORK OF LIES ???? Watch the video below again and realize that Intel CANNOT COMPETE with AMD Ryzen 4000 mobile processors at any given cross over level. That the midrange AMD mobile unit can kick the ass of a full bore i9 Intel processor running in a specialty cases that has TWO FULL SIZED DESKTOP CASE FANS TUCKED INTO THE LAPTOP CASE (yes you have to slide the laptop case open laterally to expose the two giant fans and you have to hit a special turbo switch to spin the fans up and get the thing running up at full speed). https://www.youtube.com/watch?v=ZYqG31V4qtA ...... it is a YouTube, so click on it ================================================== Let's be charitable and simply say that these are all just Tiger Lake PRE-PRODUCTION teething problems. But this litany gives you a hint why the full production of introduction laptop units have now been delayed until October-November. Some of these issues need fixing before you go make 10's of thousands of machines that you will then need to perform expensive reworks upon ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/10/20 at 01:43:38 https://tekdeeps.com/why-inteltiger-lake-is-disrupting-the-notebook-market/ Source: ITWorld Korea by www.itworld.co.kr. *The article has been translated based on the content of ITWorld Korea by www.itworld.co.kr. If there is any problem regarding the content, copyright, please leave a report below the article. We will try to process as quickly as possible to protect the rights of the author. Thank you very much! http://https://www.itworld.co.kr/files/itworld/2020/09_01/story-tiger-lake-speeds-and-feeds-power-100856490-large.jpg “The TDP is the same on the specs even though the system is different, but the actual performance can vary greatly depending on cooling, OEM tuning, Intel partnership, target users and system use cases, and workloads,” Schrout said. We don’t think one design is the best, and we see that different designs may have different focus. Product design is ultimately the manufacturer’s choice.” The problem is that consumers and notebook manufacturers have different interests. Laptop makers get manufacturing flexibility, but consumers can’t know whether the laptop they’re buying will run at a lower or higher performance state. “Tiger Lake’s power lies in its scalability and dynamic diversity of products,” Schlaut said. We do not enforce branding, thermal design, or performance delivery design. Therefore, it is only possible to judge how well the performance is through performance reviews and explanations.” http://https://www.itworld.co.kr/files/itworld/2020/09_01/story-intel-power-1-100856489-large.jpg This chart is showing the same processor applied to different functional scenarios (always limited by the chipset exceeding the cooling capacity of the laptop). According to Schlaut, Intel has no plans to classify the chip by its operating power state. “With the Core i3/i5/i7/i9 brand, we’ve launched the ’Evo’ brand for premium notebooks, letting consumers know about the performance they can expect from these product lines. Consumers and reviewers have a variety of software tools. We can use it to freely decide the power setting of our laptop, and we hope that our customers are interested in these advantages. After all, performance is the most important.” So, Intel says EACH Tiger Lake processor installations must be treated as an individual vis a vie the cooling system of the individual laptop. Intel can indeed do some benchmark tests at up to 80 watts (for extremely short periods of time) causing the chipset to throttle way down to 5 watts during a protracted thermal recovery period. This is considered by Intel to be "normal usage" not hyper clocking (cheating out their arse) as everybody else sees it. You (as the purchaser) do not get to know what you are buying any longer, except that it falls somewhere inside the considerably broad ranges that are now advertised. Note please: any sub-optimal members of a run of Tiger Lake chipsets are still considered "within spec" for certain use ranges in other words ........ :P So what if a Tiger Lake is missing a completely dead core or has two "damaged" cores, that's OK --- it still fits inside the UP-4 7-15 watts classification just as it is ...... RECOMMENDATION: AVOID TIGER LAKE PROCESSORS ALONG WITH THE REST OF INTEL 10nm++ PROCESSOR EQUIPPED PRODUCTS. ALSO AVOID THE EVO BRAND DESIGNATION, AS THE AI TRICKS AND THE HYPERCLOCKING TRICKS ARE INCLUDED IN THE EVO DESIGNATION. DO: Buy the latest gen AMD laptop products for the best, most honest and the most knowable and the most stable laptop and desktop performance --- and save yourself a chunk of money too. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/10/20 at 16:35:55 https://www.youtube.com/watch?v=Cp3xW4uncbk it is not a long video ...... just watch it Back in May of this year, some people who remember the several times in the past when Intel got just plain "nasty acting" during a short period of time when they WERE NOT COMPETITIVE and Intel had actually let their black bag boys go off on a little rampage. Intel was specifically warned in May against the sort of bad black bag stuff that is currently going on with Tiger Lake introduction. Earlier this year (in May) Intel got WARNED not to do this sort of stuff since Intel got so non-competitive early this year and it was sensed that another black bag incursion was coming as the massive angry flashback from the industry on a TOTAL LOSS OF TRUST on two past occasions was SO severe that it took Intel literally YEARS and YEARS of proper behavior (and a management change) to get that trust back again. Intel still hasn't recovered their image from the huge freon cooled mainframe motherboard that was actually running the test software down in the soundproofed enclosure that was hidden under the skirted display table during the CES show 2 years ago. :P Do you think it strange that pundits are actually WARNING Intel right now to restrain their black bag boys for Intel's own good? ================================================== So far to be a Tiger Lake "winner" for Intel you have to have a laptop with excellent heat dispersion construction, a fully functional FAST Tiger Lake chipset (some aren't so completely fully functional NOR are they all specially selected to be EVO speed level fast). You will also need a really really fast internet connection to allow the current as of today Intel tweeked libraries fast and full access during the testing runs and most importantly you need to only use the exact copies of the exact benchmarks that Intel "approves" (the ones that they have tweeked with to interface with the libraries to enable a lot of biased AI secret usage (used just during the testing of the Intel processor of course, but definitely not during the competition's test runs). Watch out for gross hyperclocking during the Intel test runs ....... i.e. watch out for several seconds of 80 watt power draw (note that your oversized laptop power supply has to be always plugged into the machine if you want Intel to really shine test after test, or you can just run it off the battery if you want a little more FAIRNESS injected into the mixture. Intel is already saying "You can't accurately predict the performance of a Tiger Lake laptop as it changes according to thermals reported from the motherboard" to give them the big out they will start needing in a couple of weeks. There will be more niggling little items discovered while the reviewers are actually running these comparisons and the pundits will quickly show us which pundits are in Intel's pocket and which ones are not. Tame Intel paid reviewers are pushing the company sponsored viewpoint that Intel wants you indoctrinated to accept ----- that you Can Never Really Know what you have bought because Intel Tiger Lake always reacts differently to different situations. They want you to BELIEVE Tiger Lake is always superior and to think that "benchmarking" is obsolete in the Tiger Lake era. :P If you are mullet enough to buy into this thinking, then you deserve the Intel processor that you just paid MORE FOR, and are going to get less from. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/11/20 at 11:01:41 https://liliputing.com/2020/09/acer-revives-the-gateway-brand-for-a-line-of-walmart-exclusive-laptops-and-tablets.html Do you remember the cow boxes from Gateway Computing? I do, I bought one. It was my first 64 bit Athlon 1 ghz single core AMD desktop chipset from way back in the day. Well, what was old is new again. Acer who bought the corpse of Gateway Computing has revived the name as their Walmart Exclusive Only brand name. So , say goodby to Onn (who did not do well against Amazon Fire tablets for features or price) and say hello to Gateway Computing ....... all coming to your local Walmart by Christmas 2020 with some having already arrived. Here’s a run-down of the Acer/Gateway/Walmart launch lineup. Note that some models aren’t yet available for purchase, and while the MSRP prices are listed, some models are already on sale for as much as $300 off the list price. Android Tablets 8[ch8243] Android Tablet (MSRP: $70) 800X1280 IPS, A50, 1GB/32GB 10.1[ch8243] Android Tablet (MSRP: $80) 800X1280 IPS, A50, 1GB/32GB Windows Laptops 11.6[ch8243] Ultra Slim Laptop (MSRP: $200) w/AMD A4-9120e, 4GB / 64GB 11.6[ch8243] Windows 2-in-1 Laptop (MSRP: $300) w/Intel Celeron N3350, 4GB / 64GB (Includes 1 year Microsoft 365 Personal) 14.1[ch8243] Ultra Slim Laptop (MSRP: $240) w/Intel Celeron, 4GB / 64GB (Includes 1 ysome models are already on sale for as much as $300 off the list priceear Microsoft 365 Personal) 14.1[ch8243] Ultra Slim Laptop (MSRP: $430) w/AMD Ryzen 3 3200U, 4GB / 128GB 14.1[ch8243] Ultra Slim Laptop (MSRP: $460) w/Intel Core i3-1005G1, 4GB / 128GB 14.1[ch8243] Ultra Slim Laptop (MSRP: $600) w/Intel Core i5, 16GB/ 256GB 15.6[ch8243] Ultra Slim Laptop (MSRP: $650) w/Intel Core i5-1035G1, 16GB / 256GB 15.6[ch8243] Laptop (MSRP: $900) w/AMD Ryzen 5 4600H / 8GB / 256GB, NVIDIA 1650 GTX GPU 15.6[ch8243] Creators Series Laptop (MSRP: $1200) w/Intel Core i5, 8GB / 256GB, NVIDIA 2060 RTX GPU Just for an example so you can see one https://gatewayusa.com/GWAT10-1.html http://https://gatewayusa.com/images/products/gwat10-1.png |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/11/20 at 22:15:53 ;D ....... yeah, folks are still getting sorta hot at Intel for their Black Bag BS ........ https://www.youtube.com/watch?v=vuaiqcjf0bs&t=435s Why show this again? It is the first time Intel attempted to ring in tuned AI usage as an illicit "processor performance boost" and reported it as "processor advancements". Ring it forward 6 months and you can see the current Tiger Lake Black Bag tricks shown in light of their origins. When this video was filmed, they did not understand completely what Intel was doing with AI, just that this processor performed EXACTLY like the last one in lots of tests, but showed better results on the AI tweeked benchmarks that were approved by Intel. It was a curiosity and a puzzlement ..... Intel cheating was suspected, but since AMD Threadripper so totally overwrapped it in the next few weeks the entire thing went into the "past history, water over the dam" trash bin. I predict that the first real production unit shipments of Tiger Lake will make up a nice "tear Intel a new one" moment from the non-Intel computer press and then will be quickly swept off the stage by Intel's backing off the Black Bag Boys bogus pronouncements and finally by AMD finally getting irritated and releasing a REAL product to "threadripper" it with lots of real raw performance that Intel can't cheat enough to overcome. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/13/20 at 20:41:19 https://liliputing.com/2020/09/nvidia-is-buying-arm-for-40-billion.html Four years after buying Arm Holdings for $32 billion, Softbank is selling Arm to NVIDIA for $40 billion. It’s the most expensive deal in the semiconductor space to date, and it would put NVIDIA in control of the technology that powers most of the world’s smartphones and a growing number of other devices. The companies say it will likely take about a year and a half for the acquisition to be completed, because, among other things, it will need to be approved by regulators in the UK, US, China, and the European Union. It’ll be interesting to see what those regulators have to say. Technically Arm and NVIDIA are not direct competitors, since Arm licenses chip designs, but does not make its own processors, while NVIDIA designs and produces its own chips. But if the deal goes through, NVIDIA will be both competing with chip companies like Qualcomm, and licensing its designs to them. And if the deal does pass muster with regulators, will chip makers that are currently licensing Arm designs decide to look elsewhere? So far pundits say that under Nvidia's control ARM will either become very very pushy in PC and in mainframe space or else they will see major customers defect to RISC-V because they don't want to play in Nvidia space under ever changing rules set by Jen-Hsun "Jensen" Huang. Odds of final regulator approval by every regulating body of the deal is less than 50% right now according to industry pundits ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/15/20 at 07:07:14 https://www.idgconnect.com/news/1506091/intel-coyly-revealed-core-tiger-lake-cpu-fight-amds-ryzen-4000 http://https://image.chitra.live/api/v1/wps/fb666fe/bb9e4e0f-d39a-41ac-bea4-2a867a0526ce/0/tiger-100856437-orig-620x354.jpg This is not a small laptop processor, Intel Boyd Phelps you little sneaky poo, you ...... this thing is HUGE. That silver edged huge thing held down by 3 screws is actually a processor sub-mother board that holds your Intel Processor daughter board and the 21 additional I/Cs (plus hundreds of other components) and all of the cache memory chips and I/O functions that AMD puts in their chipset itself (by doing so means AMD is simply not requiring much of the additional I/Cs and the other minor pieces of silicon). Can you say "Wow, what a huge oversized EXPENSIVE abortion"? Takes a lot of room, doesn't it? Just look at all of it ----- and realize that AMD puts all of that on top of the CPU die as an integral part of the processor using the four additional etch layers that ASML lithography equipment can currently lay down at 7nm. And next year at 5nm AMD's new AM5 socket will hold brand new 5nm chipsets with up to 14 additional interconnected layers laid on top of the primary silicon. I betcha Intel tries to claim this abortion is an "upgradeable processor board" and tries to announce it as another "industry first" in laptop space --- until someone reminds them that simply changing out an a AMD multi-level processor on its AM4 socket accomplishes the same thing at far far lower cost. BUSINESS MANAGEMENT Intel coyly revealed an 8-core Tiger Lake CPU to fight AMD's Ryzen 4000 Posted by Gordon Mah Ungon September 13 2020 Intel has quietly confirmed that it will do an 8-core version of its 11th-gen Tiger Lake chip. In a September 3 blog post on Medium noticed by Tom’s Hardware, Intel’s Boyd Phelps confirmed that in addition to the quad-core 11th-gen CPUs due in October, the company will also push out an 8-core variant. “The Willow Cove core increases the mid-level cache to 1.25MB—up from 512KB,” Phelps wrote, on the site usually associated with 12,000-word essays on avocado toast. “We also added a 3MB non-inclusive last-level-cache (LLC) per core slice. A single core workload has access to 12MB of LLC in the 4-core die or up to 24MB in the 8-core die configuration (more detail on 8-core products at a later date).” Intel has so far only confirmed quad-core versions using what it now calls “UP3” and “UP4” designations. How the 8-core version shakes out isn’t known. Based on the deluge of Tiger Lake laptops pushed out two weeks ago, we think the approach could be similar to AMD's. The "the deluge of Tiger Lake laptops pushed out two weeks ago" has been postponed until October/November due to unspecified delivery issues. The months long delays imposed on the first actual competitive laptops and this somewhat rushed new product announcement tends to indicate the "TOTAL VICTORY" of Intel's Tiger Lake was really somewhat less than total. If four Tiger Lake cores had overheating and battery draining problems, what do you think 8 Tiger Lake cores will do? Soon, we shall see ....... The total back handedness of this announcement is what is so striking about it. Intel’s Boyd Phelps confirmed that in addition to the quad-core 11th-gen CPUs due in October, the company will also push out an 8-core variant. “The Willow Cove core increases the mid-level cache to 1.25MB—up from 512KB,” Phelps wrote, on the site usually associated with 12,000-word essays on avocado toast. “We also added a 3MB non-inclusive last-level-cache (LLC) per core slice. A single core workload has access to 12MB of LLC in the 4-core die or up to 24MB in the 8-core die configuration (more detail on 8-core products at a later date).” How else can Intel admit that the BS they announced at first was simply "dribbling down their chins" PR BS ...... but hey, let's jest sneak the corrections in where the avocado toast 12,000 word essays normally go. Now, Intel, how do you go about RETRACTING all that "core count equivalency" BS that you frothed out about during your main Tiger Lake introduction ????? You would only compare Tiger Lake to much older AMD stuff that had only four cores in it, remember? Or, is this Intel's way of finally admitting they really don't currently REALLY HAVE a competitive product to sell? :P Duh, Intel ..... and who is copying who now again ????? Level 4 cache copied, large bulk systems memory copied, total core counts copied, hey they say you copied other things as well but that has not been verified at this point. So, how do you think Intel will do against the newer AMD products that will have come out by the time you can limp out your physically HUGE overly expensive 8 core thermal killers? :-/ You do know that you have really pissed off all the real reviewers out there, and you also pissed off Lisa Su as well with all your baseless assertions? Given your slow reaction time, I think everybody will be all ready for you when you finally send something out for them to get their hot little hands on ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/15/20 at 14:12:09 https://liliputing.com/2020/09/sifive-plans-to-unveil-the-first-risc-v-pc-in-october-along-with-new-risc-v-chips.html https://www.cnx-software.com/2020/09/15/sifive-to-debut-risc-v-pc-for-developers-based-on-freedom-u740-next-gen-soc/ http://https://liliputing.com/wp-content/uploads/2020/09/sifive3.jpg A company called SiFive has been producing RISC-V chips for the past few years, and now the company has announced plans to introduce the “world’s first RISC-V PC” during the Linley Group Fall Virtual Processor Conference at the end of October. Aimed at developers rather than end users, the idea is to give people a computer that they can use to code software for RISC-V powered devices while using a RISC-V device. At its heart, the upcoming computer will feature a SiFive FU740 processor, which is a new system-on-a-chip featuring a “heterogeneous mix+match core complex” with SiFive U7-series CPU cores, as well as support for PC form factors and expansion capabilities. SiFive says it also plans to introduce a new processor core optimized for “high-performance computing, artificial intelligence, and computer vision applications.” Interestingly, those are exactly the same sort of applications NVIDIA highlighted as a key reason that the company is purchasing ARM. ::) I think ARM is under some very early light small arms fire by this one, which may be one of the reasons why Softbank was willing to sell ARM off to Nvidia ....... Hey Intel, a little gift for you is coming this fall about the same time you finally ship some Tiger Lake laptops ...... yet another up and coming competitor for both you and ARM and Nvidia is now taking its very first PC baby steps. Note please Intel, that the entire HiFive motherboard is a whole lot smaller than the Tiger Lake processor sub-board on the Tiger Lake thingie that you just announced. And I also note existing unused layout positions on the HiFive board for 5 more memory modules that are already tracer wired into the board itself ....... Got plans, HiFive does. This is the processor layout for a 9 core RISC-5 processor, which would be the VERY LEAST of the PC capable RISC-5 processors HiFive is building next year. http://https://linuxgizmos.com/files/sifive_u8_soc.jpg ================================================== This upcoming PC is not intended for consumers to use as a PC per se ---- it is a test bench for Linux programmers to use while writing Risc-V code as debugging freshly written code is always best done native to the environment you are coding for. As long as standards are maintained, if it runs after debugging during development it will run in consumer use across the board. ================================================== MIPS has just been bought again by the Chinese --- I bet they wish they had kept it the first time they bought it as they paid more for it this time around. Why buy it back? Code and hardware tricks that you steal from ARM can be tweeked to run on MIPS and by this "backhanded excuse" Chinese phones and PCs can stay relatively more current technologically during the Trump embargos. And nobody wants to be 100% under Jensen Huang's thumb when the ARM deal closes ..... My theory is that China actually kept the tech even after selling it as their military uses MIPS based mil-spec machines built specifically for them in a Chinese MIPS production facility (using stolen Los Alamos labs EMP hardened MIPS design tech from the '60s). Now China has a legal right to what they stole in the past. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/16/20 at 10:01:33 https://liliputing.com/2020/09/intel-quietly-launches-two-new-comet-lake-h-processors.html Intel quietly launches two new Comet Lake-H processors as backup for delayed 10nm launches. When Intel launched its Comet Lake H family of 45-watt chips for laptops and compact desktop computers in April, there were six processors in the lineup. Now there are nine. A little while back, Intel added the Core i9-10885H, and this week Intel quietly added two new Comet Lake-H processors to its website, and according to a recent Reddit post from the makers of the XMG line of gaming laptops, the reason may be due to supply issues with some of Intel’s other chips. Root behind this issuance of old 14nm designs is the much delayed 10nm ++ generation chipsets and also delays making the H series 14nm chipsets that were supposed to back up the 10nm ++ products. Delays in Tiger Lake also play into this decision to reissue old 14nm designs. Apparently Intel is having trouble cranking out enough Core i7-10875H processors, so the company has decided to produce a slightly less powerful Core i7-10870H to help meet demand. It’s base and turbo clock speeds are 100 MHz lower, but otherwise it’s pretty much the same 8-core, 16-thread chip with 1.2 GHz Intel UHD graphics, 16MB of cache, and support for up to 128GB of DDR4-2933 memory. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/16/20 at 10:14:43 Rumor Time again. It is rumored that Intel now has to pay up front now for any production runs at TSMC or at Samsung. The rumored reason is that the Intel designs don't always work as contracted to be built, and nobody else is interested in holding the bag for a failed Intel production run yet again. And it is rumored that Intel likes to stick their partners (Micron for example) for the dead warehouse stock from their own Intel goof ups if they can. And is it rumored that just now that a run or two of Tiger Lake has just bit the dust due to inherent design issues (it was developed at Samsung and then was actually run at TSMC because it was cheaper --- and Intel lost track of the process differences at the two foundries) Veracity of this rumor is unknown but this one should become apparent as real or not by Christmas time. =================================================== Another thing that is rumored to have happened whenever Intel hits a road bump with a new processor ...... Some of their builder/partners will have early advertised units with the new lithography processor well in advance and have taken lots of orders for that unit. Intel cannot deliver and falls back on a alternate processor that isn't quite as good. If the builder wants to build his units (for Christmas just to grab an example out of the air) he has to take the alternate lot of Intel processors or accept a sort level that is way short of the original specifications ....... The builder then has to push the replacement processor built units out the door on to the end seller (generally a retailer chain of some sort) as a modified part number with a letter tagged on to the end of the part number. In turn the retailer has to have a reduced price sale to move out the substandard units (during which he conveniently sometimes loses track of reporting the substandard build characteristics to everybody involved) and it gets out into the public channels as the original advertised product and is touted by somebody's out of the loop sales department as "meeting the original specs". :o With Intel, this is rumored to have happened ...... distressingly fairly regularly whenever Intel has to use a series of backup processors to make their ship commitments on time. =================================================== Whenever you see Intel suddenly widening up the performance range of a new chipset (as they just did with 10nm ++ Tiger Lake) you can kinda count on another confused mess is coming on down the pike. Especially if SORTING and BINNING gets mentioned in passing ...... Especially if Intel is suddenly all of the opinion that the performance of the new Tiger Lake 10nm ++ thingie "cannot be determined by existing benchmarks". :o :P (sound familiar to you at all?) Then when the backup 14nm chipset for the 10nm ++ that is having the problems gets an older 14nm backup suddenly pulled out of warehouse mothball stock (a less capable, older chipset) as its very own special backup due to simply not being able to make or deliver enough of first backup chipset ---- well then you can suspect multiple entire series of lots of chipsets are perhaps involved in subtitutions about now and it is perhaps rumored that a right snarly mess of some size has taken place that is being kept a secret by Intel and Intel's builders just so they can have something to sell this Christmas. ::) hmmmmm ?? Unsubstantiated rumors ......... don't you jest love them? They compound on each other and dovetail together so nicely. What imaginations these rumor people must have to think Intel would actually do such shenanigans ..... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/17/20 at 08:56:38 https://liliputing.com/2020/09/3rd-party-intel-tiger-lake-tests-suggest-big-performance-boosts.html http://https://liliputing.com/wp-content/uploads/2020/09/tlake_04.jpg Intel is really stumbling over their lack of shipment of real normal channel production based competitive Tiger Lake laptops -- and Intel is now making an effort to CUSTOM SUPPLY a laptop to AnandTech, Engadget, Gizmodo, PC Magazine, The Verge, and Tom’s Hardware --- this custom unit is a reference design laptop that is provided by Intel that is NON-STANDARD in several ways. First, this is not an actual product you can go out and buy. It likely has a cherry picked processor and faster than normal memory installed in it. Second, The reference unit likely has a cherry picked CPU out of Intel’s best runs of the most powerful 11th-gen chip – the Intel Core i7-1185G7. It’s unclear if we’ll see the same kind of performance improvements from a standard run of the mill chipset of the same model and type from Intel or from TSMC ---- much less a run of the mill Core i3, Core i5, or even other Core i7 chips in the real upcoming Tiger Lake family. Finally, this likely "cherry picked chip" has a configurable TDP, allowing it to run at 12 to 28 watts "by direction". Ice Lake chips only go up to 25 watts, and most laptops with this processor are shipped with a 15W configuration to maximize battery life. AMD’s Ryzen 4000 series notebooks also top out at 25 watts and, again, many laptops that ship with these chips are configured to run at 15 watts to maximize battery life. The Intel reference unit is loaded up with all sorts of AI software speedups, that will automatically play in all the testing done on this machine. The reference unit also does not seem to throttle very much as noted during initial limited testing, which is not normal at all for a laptop as the processor heats the chassis up. This "special effort" by Intel comes across as a whole new layer of attempted PR manipulation and most of the test houses simply state their intention to wait for a real test unit that they buy out of the normal retail chain to base their tests results upon. One of the group states they intend to do the tests with the machine that they buy, then they intend to replace and format the hard drive of their purchased machine and install an older version of Windows Pro that has no AI enhancements in it (nor would it be able to ask the internet to download any of such) as they are curious how much of Intel's progress is AI manipulation vs real performance increases (real progress). Plus their own purchased laptop can be equipped with thermocouples and recording watt meters to see if any 80 watt limited time hyperclocking takes place during any of the tests as they are being run. Nobody trusts Intel not to cheat any more --- it is too important to Intel that Intel get a win with Tiger Lake any way they can ...... This last bit is quite interesting to note. "A Lenovo Slim 7 laptop with a 25 watt Ryzen 7 4800U processor does come out ahead in some tests, including those that take advantage of multithreading (it’s an 8-core/16-thread chip, while Intel’s is a 4-core/8-thread processor), and video transcoding (using Handbrake, at least)." |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/18/20 at 01:52:24 This is me talking, making predictions based upon what I have read in the last 3 months. Intel planned out their move to 10nm years and years and years ago, but now that it has finally arrived (sort of) it really isn't a super innovative thing compared to the original AMD moves to 7nm TSMC processors, much less the just being introduced 6nm and 5nm and 4nm and 3nm TSMC products which is what the entire world is getting ready to drop down to. To the market place, 10nm is a day late and a dollar short, in other words. 7nm isn't going to gain much more additional steam going forward what with 6nm being announced as the new TSMC node of choice, so the existing the existing early TSMC 7nm lines will get converted one by one to the newer below 6nm lithography levels by TSMC to keep being useful (if at all possible). Intel, what will Intel do ??? Intel will use any and all tilted AI functions and tilted background libraries and legally restrictive agreements to try to maintain their 10nm position for another 5 years, but unless they can successfully roll on down to lower lithography levels Intel will fail again. Intel will simply stay stuck at 10nm for a long long long time, way too too long just like they did at 14nm. Meanwhile, competitors using smaller, newer and more modern lithography types (5nm 14 layer immediately comes to mind) and will continue to greatly outsrip Intel in real performance. Intel is spending a lot of money and time to "muddle up and confuse" all forms of benchmarking right now trying to sell folks on the thought that using AI to speed up certain benchmark functions constitutes "processor advancements" when in fact it is simply using a software trick to seem to speed up the benchmark's reported results from the same old same old types of processors. The processors don't get any faster, and normal code doesn't run any faster, just certain parts of certain benchmarks seem to execute faster. Intel is now fully acting like a black bag company, one that runs off lawyers, dirty tricks and hidden secrets and as such INTEL IS NOT WORTHY OF YOUR TRUST ANY FURTHER. If Intel says it ---- then automatically doubt it ...... you will likely be right or very close to it. AMD is still making prompt bi-annual lithography moves keeping time with APPLE and TSMC lithography improvements. TSMC just broke ground on building the buildings in a new 3nm campus, TSMC has completely built out their 5nm campus for this year's 5nm implementations and is putting the very last of the 5nm 14 layer ASML assembly lines into it. TSMC is on track and ready to go for 5nm. Intel is supposedly planning on going over to go to TSMC as their build supplier starting at 6nm. If Intel would actually start buying up some of the "converted to 6nm" old TSMC 7nm lines to replace their own ancient 14nm lines that would be an appealing enough idea (I guess) but who knows what will actually happen --- but this is Intel we are talking about after all so the odds of them doing something quick and smart needs not even be considered ........ =================================================== Nvidia buying ARM outright throws all of ARM into "future doubt" as lots of folks are now moving from ARM over to RISC-V technology just as fast as they can --- as RISC-V is royalty free technology and is not under Jensen Huang's thumb royalty-wise. This shift is well begun starting with the smallest and simplest of the I/Cs used in hard drives, modems, general electronics, etc. More complex nine core PC and phone level RISC V processor chips are coming out later this year. This could wind up being an important move for basic level phones ....... especially when the next generation of 12 and 16 core PC and phone RISC-V processors are designed and built in 2021 by an oriental consortium. (can we say Chinese, or is that just considered being insensitive, Mr. Biden?) Change, she comes ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/22/20 at 05:06:03 https://www.notebookcheck.net/Leaked-roadmap-suggests-AMD-Zen-4-desktop-CPUs-with-DDR5-support-may-only-launch-in-2022-no-relevant-AMD-CPU-launches-planned-for-2021.463331.0.html RUMORED New Confusions over AMD Roadmaps for the next 2 years The global pandemic appears to have messed up with the launch schedule this year, as the latest rumors regarding the Zen 3 CPU launch suggest a Q4 2020 launch, so the Ryzen 4000 desktop processors could land this October at best. Now, according to an insider close to GamersNexus, it looks like AMD might have pushed the Zen 4 launch to 2022 due to the current uncertain economic climate, possibly leaving the rest of 2021 with no exciting launches past Zen 3. As it stands, the roadmap shows that AMD intends to introduce DDR5 support for the desktop CPU lineup in 2022. Apparently, the lineup in question is the Ryzen 5000 based on the Zen 4 microarchitecture, and it will also come with native USB 4.0 support. As far as desktop APUs are concerned, the roadmap shows that 2022 will bring Zen3+ models supporting DDR5, as well, whereas the mobile APUs will feature on-board LPDDR5 support. Considering that the Zen 3 CPUs may get a late launch this year, it probably makes sense for AMD to postpone the Zen 4 family to early 2022, but no mobile or desktop Zen 3 APU launch for 2021 seems like a bit of a stretch. We know TSMC is opening its 5 nm production nodes this year and AMD has already booked considerable 7 nm and 5 nm production capacities for this year, so it is hard to imagine that team red will essentially skip the entire 2021 as far as consumer processor launches go. A recent leak even points out that Dell is planning to release laptops powered by Zen 3 mobile APUs next year, so that GamersNexus roadmap clearly feels inaccurate, at least when it comes to the Zen3+ APUs. AMD sees 10nm Intel as no real threat at this time, but as always AMD can always slide their newest stuff forward in time if needed. Conversely, AMD has often sandbagged part of a year when Intel had nothing new to offer. You see, AMD likes to sell off old warehouse stocks at full price when they can, as "closing stuff out" to make room for new stuff costs AMD some money in discounts, etc. Right now AMD's firm commitments to their first production lots of 5nm production at TSMC will drive their newest "foremost groundbreaking products" ...... not Intel's poor competition performance. Question remains, what will AMD use the first lots of 5nm chiplets to build first? |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/22/20 at 06:15:09 https://www.youtube.com/watch?v=Wfd3cO2gVHE Just click on it Intel actually sent Linus Tech Reviews a tweeked Intel owned proprietary review laptop, but Intel required Linus NOT TO SHOW THE LAPTOP'S INTERNALS NOR TO TEST NOR TO TALK ABOUT THE LAPTOP'S BATTERY LIFE ..... Linus does a lot of grinning while he is talking, as he goes over much of the same things that others have brought forward so far. He grins a lot when he tells you what he isn't allowed to show you and talks about those items he is not allowed test on camera and then he shows you some of the items he can't even talk about. He's genius, really. He talks about what he can't show, he shows what he can't talk about and he initiates a benchmark which he then lets runs blind on both machines where you (the viewer) can see it in the background while Linus stands off to the side with his back to the machines where he can't possibly even see what the benchmark is showing us ---- but I think he must know the results ahead of time since he talks around the images that are developing on the two machines as "possible items" without being able to see them at all. Grins a lot, really ...... he is really enjoying himself as he gives us his best impression of a smug Intel "fact dodging press review" as he does all the standard Intel misleading and deceitful press review tricks ....... By his demeanor, it is clear that Tiger Lake is possibly not A HOT RUNNING POWER HOG and this same "not review" also Linus infers that when both AMD and Intel Tiger Lake are run at the same watts draw the AMD products will win out easily (an easy guess since AMD would win anyway due to 2x the number of cores and at ANY given power draw AMD would draw far less battery to get the same job finished first (tasks that are covered by specific AI speed ups are exceptions although it is clear that AMD also gets it all done QUICKER in general anyway). Most carefully stated are Linus' impressions of Tiger Lake's graphics performance, an item in which AI can show a profound % change that really has everything to do with the AI software tweeks, not anything real concerning any real improvements in the Tiger Lake GPU or the CPU that is running the AI software. Linus then simply promises an unfettered real review later on using a real (un-tweeked) production unit purchased from independent retail sources. He then mocks Intel's words about "each laptop being a world unto itself" as far as any performance metrics go. ;D (Linus mentions all this while grinning widely) Linus also mentions that "real production laptops" generally tend to be set for a 15 watt power draw level so as to maximize battery life ...... and the inferences provided here hinted at that Intel has set this one to around 21 watts to maximize the test results before any gross chassis soaking heat throttling takes place. (Intel clearly knowing all about the tests they put on the machine for the reviewers to use in some great detail, enough to hand tune the thermal ramp down of the preview laptop anyway). |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/23/20 at 12:55:45 Short term supply issues with AMD Ryzen 4000 U series laptop chipsets are rumored to be quite severe right now as three of these AMD Ryzen 4000 U series laptop items (on 3 different brand names no less) are #1 Best Sellers right now. Retail outlets are still taking massive orders for the very efficient and low cost laptops. Back order delay times of up to 3 months are being logged in now on Amazon. SOMETHING HAS GOT TO GIVE. AMD needs to bring forward additional larger supplies of AMD Ryzen 4000 U series chipsets ASAP or they will open the door for Intel Tiger Lake to go ahead and waltz on in even though the AMD is clearly the better (and much cheaper) chipset. Intel Tiger Lake is also being delayed right now, and may not be able to move quickly enough to seize this opportunity. AMD has gotten close to getting into trouble with lack of supply a couple of times during the ramp up of Ryzen, but this is the first time it looks like a "rather large issue". AMD has always tended to give TSMC production time to more expensive products when they hit a tight patch on TSMC production time. The laptop chips are not chiplet based, but are APUs (which are still old style discrete processors). Once you are at TSMC, you are simply dealing with how you spend your production allocation in wafers, whether you want X wafers of which chiplets or you want X wafers of APUs instead ...... As said, something has got to give. One way is to go ahead and move on down to 5nm with more expensive chiplet based products to free up your old 7nm allocation for more APUs ..... this is planned for next year, but the issue is NOW, this year. This impressive bump in AMD Ryzen 4000 U series chipset demand has only one visible reason --- Intel has highlighted just how good the AMD units are by all their cheating shenanigans they played with the Tiger Lake introduction. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/24/20 at 20:59:48 https://youtu.be/i2jFN_QEcS4 it is a youtube, jest click on it ..... a flying camera drone from AMAZON, just what every young male teenage gadgeteer needs to get into even more and deeper trouble ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/27/20 at 02:46:44 https://arstechnica.com/gadgets/2020/09/hands-on-with-intels-i7-1185g7-tiger-lake-prototype-laptop/ http://https://cdn.arstechnica.net/wp-content/uploads/2020/09/Hero-Shot-800x600.jpg Arstechnica.com is a FOSS Linux site that primarily does Linux reviews, and as such the fact that Intel sent them a Intel built and owned Tiger Lake prototype just to see if they would heap kudos on Intel's head over Tiger Lake out of pure ignorance was another incorrect assumption that was made at Intel Marketing. Wrong folks, Intel, they are not dumbies over at ARS --- you got a meticulously honest knowledgeable review, which we tend to think really wasn't what you were really after. A VERY HONEST review ...... And you apparently either forgot to tell Arstechnica all of your little fiddly little review rules (or else being a Linux site they "free and open source" plain IGNORED some of your restrictions). Time, RUGs, and relevance We regretfully did not spend much time on Intel's provided RUGs (Reference Usage Guidelines), which rely on real-world applications such as Adobe Photoshop and Premiere, Topaz Gigapixel AI, and Nero AI Photo Tagger. While we do agree with Intel that real-life workloads are critically important, it's difficult to separate Intel's own search for favorable comparisons from the specific features the company chooses to test inside those real world applications. ::) In particular, any applications running AI inference workloads using OpenVINO will benefit enormously from Intel's AVX-512 instructions (the company brands this as Deep Learning Boost). This is a real advantage for Intel, in those applications, and we expect this advantage will increase over time as more applications begin integrating inference workflows. You can already see such workflows in applications like Photoshop for some filters and "magic select" tools. Ars didn't do anything with the list of "approved" Intel tests as Ars saw them as pure BS PR management on Intel's behalf. Ars also does not think that AI trickery has much to do with processor performance. Bullet points lifted from the rather long main body of the review Hyper-clocking up to 68 watts was detected by ARS The system also exhibited noticeable capacitor whine. If you have a good ear for electronics, you can actually hear the device "thinking" from the change in the faint, high-pitched noise as the CPU shifts in and out of turbo frequencies, which it does rapidly. We also did not test and will not talk about the system's battery life in this review. Again, this isn't really a laptop to be reviewed—it's just a platform that enables us to put the i7-1185G7 to the test. We did, however, check the system power consumption at the wall using a Kill-A-Watt meter. Desktop idle varies from 7.5W-8.2W, and peak consumption (as measured during the first several seconds of a Cinebench R20 run) is about 66W-68W. Any Tiger Lake i7-1185G7 system can briefly reach maximum turbo frequency and hit a power-consumption max of upwards of 50W, but the system doesn't stay there for more than a few seconds, before dropping clock enough to fall back to its configured TDP. The length of time which the system is allowed to stay at the highest hyper-clocking consumption (PL2) is called “tau.” Tau, along with the TDP itself, is configurable by the OEM. We seriously doubt tau will be disclosed on the box or in the advertising copy for many systems—and it can potentially have an even bigger impact than the TDP. A laptop allowed to run at PL2 (and draw 50+W) for minutes at a time will perform considerably higher (and exhibit a hellaciously higher power draw) than a competing system with the same TDP but a tau of only a few seconds. In its default configuration, with a 28W TDP and roughly 25-second (65-68 watt draw during tau) the Tiger Lake i7-1185G7 reference system runs pretty much neck-and-neck with a Ryzen 7 4700U-equipped Swift 3. We see the same basic relationship among all systems on both Cinebench R20 and Passmark. Geekbench 5 flattens the differences between all the processors in general, but it puts the 1185G7 at the head of the pack. However, we suspect many retail 1185G7 systems won't be configured for a 28W TDP—they'll likely be configured at 15W TDP, just as both the Ryzen 7 4700U in the Swift 3 and the Ice Lake i7-1065G7 in the Dell XPS 13 are. When we drop the reference-system TDP limit to 15W by moving the slider hard left, we see a significant drop in performance. It actually falls slightly below the Dell's Ice Lake 1065G7 on Cinebench R20, let alone the Acer's Ryzen 7 4700U. Attempting to compare top of the line to top of the line (saying the tweeked & cheating in multiple manners Intel owned Tiger Lake unit is a synthetic top of the line is OK enough, we guess) doing this it is possible to get somewhat close to a unit to unit comparison. Since the i7-1185G7 is the highest-performance SKU announced for Tiger Lake, we felt that comparing it to the Ryzen 9 4900HS in our Asus ROG 14 gaming laptop would only be fair. The 4900HS in the ROG 14 is running at 35W TDP, and it utterly dominates the Tiger Lake i7 in both Cinebench R20 and Passmark testing. This is without any hyperclocking being used on the AMD unit and yeilding the standard Asus ROG 14 battery life ...... Conclusions Tiger Lake as an exciting improvement from Ice Lake, Comet Lake, and prior architectures, which finally makes it a somewhat reasonable competitor for AMD's Renoir. Or — just as accurately — _you can scoff at the Intel Tiger Lake's need to nearly double Renoir's power consumption before the highest-spec Tiger Lake CPU can barely manage to compete "roughly evenly" with a middling level Renoir on most tests. If your workload is unusually (very predominately) AI-dependent, Tiger Lake is likely a slam dunk for you. Its AVX-512 instruction set gives it an enormous advantage over otherwise more powerful and efficient AMD systems. But we need to stress the word unusually: inference workloads are more common and important every year, but they shouldn't be most people's deciding factor yet. Where Tiger Lake shines brightest is in the narrow realm of laptop gaming without a discrete GPU. Although the CPU portion of Tiger Lake has difficulty holding its own against AMD's Renoir, its new Xe graphics look to be the new gold standard for iGPUs. Xe isn't ready to compete with high-end discrete GPUs in the Nvidia RTX series, but we suspect it's rung the death knell for Nvidia's lower-performance, more power-efficient MX line. The final question for any new Intel CPU line in 2020, of course, is what availability will be like. But since multiple vendors such as MSI and Acer have already announced entire product lines based on Tiger Lake, we don't expect to see a repeat of last year's Ice Lake unobtainium fiasco. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/27/20 at 03:57:13 https://www.anandtech.com/show/16084/intel-tiger-lake-review-deep-dive-core-11th-gen/6 Parts of this article were referenced in the post above this, so in fairness I list the charts and give the source so you can take a good look at it. Intel Tiger Lake always hyper-clocks at the start of every large work load until the chipset gets hot enough to hit a thermal limit, then Tiger Lake has several different recovery curves depending on the way the laptop is set up by the builder. http://https://images.anandtech.com/doci/16084/Power%20-%20TGL%20vs%20TGL%20vs%20TGL%20yCruncher_575px.png This consistent Intel Tiger Lake "feature" goes a long way towards explaining the "hot running" and "power hog" designations the Tiger Lake family has been rewarded with so far. :P The following chart monitors hyper-clocking behaviors while running an Intel approved test suite, one with strong AI usage built into it --- you can see the hyper-clocking being triggered inside an AI test run which smells strongly of "jest plain old fashioned pre-programmed test cheating" to me. http://https://images.anandtech.com/doci/16084/Power%20-%20TGL%20vs%20TGL%20Agisoft_575px.png NOTING THE LONG RECOVERY PERIODS SHOWN INSIDE THE BLUE LINE'S HYPER-CLOCKING SECTIONS AT THE END --- I really have to question if Tiger Lake's hyper-clocking on AI trick really yields all that much in in "additional processing power" for any very real types of very short period day-to-day tasks. You will trigger the hyper-clocking repeatedly task by task (it comes up hard and then after about 25 seconds of full load you have to pay the 50-100 second "slowdown" recovery period for each short hyper-threading as the hyper-clocking function ends) -- this could wind up costing you more in overall processing time than the average time savings that the short duration hyper-clocking on AI trick actually gives you on the smaller tasks. ::) And it will certainly overheat your laptop and eat your battery life ....... Face it, we don't generally do many things on a common laptop that take 3,000-4,000 seconds to do, period. Much less an AI rich task set that takes 4,000 seconds. However we can do a lot of 30 second or slightly longer tasks that will require more in recovery seconds than they save in the first place. And if it takes 4,000 seconds worth of AI rich tasks to show a good advantage for Tiger Lake, then hey, mebbe Tiger Lake ain't such a great deal for normal people after all. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/27/20 at 15:47:27 https://www.techpowerup.com/268650/amd-confirms-vermeer-zen-3-not-delayed-to-launch-in-2020 AMD fans were quiet until the main competitive reviews had landed (the ones from Tom's Hardware, Anandtech and Arstechnica and others -- the very well respected very good technical reviewers in other words). These reviews have now landed and exposed Intel's dirty tricks for what they are ----- bogus PR manipulations that are in the same spirit as Intel sneaking a mainframe motherboard with a freon refrigerant cooler into a soundproof enclosure under the skirted display table at CES three years ago. But the AMD fan base are now getting some licks of their own in, in a somewhat pointed fashion, to counter all the somewhat baseless Intel Marketing propaganda that has been put out by Intel in the past 4 months. http://https://memegenerator.net/img/instances/81059761/if-you-want-me-to-take-a-dump-in-a-box-and-mark-it-intel-i-will-ive-got-spare-time.jpg Kinda strong wording there, boys, but we have all read the rumors about 12 core and 16 core 5nm lithography at 14 layers AMD AM5 socketed laptop products with the special AI processor layers as part of the center chipset so AMD can run the bogus Intel benchmarks even better and faster than Intel can (and at a much much lower power draw, too). Poetic justice, that's what that would be. Betcha AMD can actually build theirs quicker than Intel can build their proposed 8 core Tiger Lake from scratch ..... ;D Go ahead and piss off Lisa Su you stupid Intel idiots, I dare you ....... ==================================================== FOR REFERENCE FROM 3 YEARS AGO ....... https://www.tomshardware.com/news/intel-28-core-cpu-5ghz,37244.html https://www.tomshardware.com/news/intel-28-core-processor-5ghz-motherboard,37213.html http://https://cdn.mos.cms.futurecdn.net/ZArMthqWeP23gGyZxZurL7-650-80.jpg http://https://cdn.mos.cms.futurecdn.net/eMPhYW2AxBn9qj8sgv6EsT-650-80.jpg http://https://cdn.mos.cms.futurecdn.net/tkAiCENfeWLkJCY7ZRMrRa-650-80.jpg.we Just re-read the articles and note that "Intel forgot to mention" is still a favorite Intel BS excuse that Intel still uses today ...... https://www.tomshardware.com/news/intel-28-core-cpu-5ghz,37244.html https://www.tomshardware.com/news/intel-28-core-processor-5ghz-motherboard,37213.html |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/29/20 at 00:43:51 https://liliputing.com/2020/09/review-acer-swift-3-with-ryzen-7-4700u-is-a-650-laptop-that-punches-above-its-class.html FRESH SALT for sprinkling on Intel's open wounds INTEL First, they get WARNED up front not to do this sort of crap ever again, then Intel goes and does it anyway ..... They get caught just last week doing it again, exposed, then debugged by 3 major review houses. Then Intel gets reminded of the last time they did this sort of crap and got caught at it and publicly embarrassed by the computer press ....... ;D and now they get caught again and will now be cheerfully pounded into the ground like a tent stake ...... So now AMAZON puts the Acer Swift 3 with Ryzen 7 4700U (a $680 laptop) on SALE for $655 !!!!! --- yep, the Tiger Lake crusher laptop is now on SALE at Amazon !!!!! Review: Acer Swift 3 with Ryzen 7 4700U is a $680 laptop that punches above its class The Acer Swift 3 line of laptops are thin, light, and designed to offer a good balance of performance, mobility, and price – and the latest version seems to very much deliver on that promise. The Acer Swift 3 (SF314-42-R9YN) is a 14 inch laptop that weighs less than 2.7 pounds, measures just over 0.6 inches thick and which has a suggested retail price of $680. But it outperforms many laptops that cost twice as much in many benchmarks. http://https://liliputing.com/wp-content/uploads/2020/09/swift-3-front_01-768x532.jpg That’s largely thanks to the AMD Ryzen 7 4700U processor that powers this laptop. It’s a 15-watt, 8-core chip that runs neck-and-neck with Intel’s Core i7-1065G7 processor in most of the performance tests I’ve run, and which can even hold it s own against more powerful chips under the right conditions. As of September 28, 2020 the best price I can find for the Acer Swift 3 with a Ryzen 7 4700U processor is $655 from Amazon. Meanwhile the Ryzen 5 4500U version is on sale from Walmart for $579. Take a real good look at this $655 laptop on sale at Amazon. It is the laptop that the early crowds of $1,000-$1,400 Tiger Lake laptops will be crushed by this fall. There will be others to follow it, as AMD has said there is no long term AMD Ryzen 7 4700U processor shortage this fall (that was just a rumor put out by the Intel Black Bag boys to encourage the buying public to pay 2x more for an Intel Tiger Lake laptop). Latest rumors indicate even more cores are planned for 5nm 15 layer from AMD, on a new pin rich AM5 socket that has provisions for more on chip memory and will run faster I/O speeds and has even greater natural throughput speed advantages (real speed, not Intel fake speed). So, settle back and watch the Intel PR machine attempt to sell you some specially produced yellow snow, made on the spot just for all you thirsty Eskimos ..... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/29/20 at 09:45:08 Intel's pet vendors promise to bring out literally a couple of dozen Tiger Lake laptops "soon" --- while loudly repeating the same list of bogus Intel claims that you now know are false exaggerations. What is sad is the following hard facts: if they run the full set of Intel cheats, including the 28 to 60 watt hyper-clocking power draws then the Intel Tiger Lake laptops will get hot, throttle deeply & repeatedly and totally suck on battery life. The resulting Tiger Lake performance level barely matches the 15 watt setting of the Acer Swift 3 with Ryzen 7 4700U that is on sale at Amazon for AMD also points out that the same Ryzen 7 4700U chipset can be thermal set by the builder to default to higher current draw numbers giving even greater performance (while still giving better thermals and battery life than Tiger Lake can give at any comparable settings). ROG Gaming for example has set their Ryzen 7 4700U gaming rig as high as 31 watts for continuous gaming uses and the AMD Ryzen 7 4700U processors apparently survives this sort of heavy load usage just fine. ROG does use a type of heavy duty thermal heat pipe solution not usually seen in normal laptops as a gaming rig is intended to run continuously at full tilt for hours on end. Laptops are thermally limited by their cooling set ups -- always. http://https://liliputing.com/wp-content/uploads/2020/09/swift-3-front_01-768x532.jpg The Acer Swift 3 with Ryzen 7 4700U laptop is currently the best selling laptop on Amazon for the last few days running --- and we shall see how long this position of sales dominance lasts. Reminder, Tiger Lake still hasn't shipped any real units yet. Period. This week the Acer Swift 3 with Ryzen 7 4700U laptop has won the PC Magazine Editor's Choice Award. Run the slider over to the left or click on the article below to see the Awards this laptop has won in the past week or two. Please remember that only bulky laptops with full sized desktop chipsets and special coolers can beat this rig, so that is what it is compared against in the article. REMEMBER THERE ARE NO REAL TIGER LAKE PRODUCTS TO COMPARE AGAINST ...... and there won't be any Tiger Lakes in any stores until Thanksgiving time. https://www.amazon.com/Acer-Octa-Core-Graphics-Fingerprint-SF314-42-R9YN/dp/B086KKKT15/ref=as_li_ss_tl?ie=UTF8&linkCode=sl1&tag=lp_daily_deals-20&linkId=773b505feb40c7ebeda94a74951bde24&language=en_US http://https://images-na.ssl-images-amazon.com/images/I/71W5ZdS%2BsEL._AC_SL1500_.jpg |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/30/20 at 15:32:42 https://www.tomshardware.com/reviews/asus-zenbook-flip-s-ux371 https://www.tomshardware.com/reviews/asus-zenbook-13-11th-gen-tiger-lake https://www.ultrabookreview.com/41503-asus-zenbook-ux393ea-review/ Here are three links to the seven first reviews of the first released Tiger Lake machines. You can clearly see which reviewers blindly followed the Intel recommended tests (yielding the expected built in bogus test results) and then there were the more experienced reviewers who did a more credible job of setting up and following "a real competitive review" strategy that showed a much more realistic picture of Tiger Lake. Really "reality testing" the machines gave results similar to this: http://https://cdn.mos.cms.futurecdn.net/3PR3JNntAAgcQZnEDLn9G9-650-80.png.we Whereas blindly following the Intel "Recommended Test Procedures" yielded this sort of results: http://https://cdn.mos.cms.futurecdn.net/PyFq9oZ8ndCTJrKHGSEZce-650-80.png.we My take on reading the first reviews was mixed, as we were dealing with some very skilled reviewers who dug in deeply and then we had some relatively shallow depth "beginner" reviewers. Some reviews just seemed to regurgitate the Intel Press Kit information which really isn't really reviewing anything, really. All of them saw some signs of "funny stuff" going on with the Tiger Lake processor's thermal characteristics during heavy use but only one reviewer got into the large amount of heat generated and dumped by his test machine using a heat sensor camera. http://https://www.ultrabookreview.com/wp-content/uploads/2020/09/asus-zenbookux393-gaming-performance-960x238.jpg One newbie reviewer just listed the Press Kit numbers for battery life and apparently never tested things herself for real to see what sort of life she really got. Others tested and noted some abnormally short battery life, only one reviewer really dug in and tried to find out why. I was amazed at how shallow some of the reviews were, and how willing some reviewers were to "stick to the Press Kit" for all their information. :P To find one of these last sort actually working for Tom's Hardware was a real disappointment, to me anyway. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/30/20 at 15:56:11 Here is a comment from a viewer looking over the same results that did, and I get behind his impressions 100%. Hifihedgehog says: 09/30/2020 at 1:36 PM This is the tell-all: “Performance sustainability is not that great as well. When running CineBench R15 xT in a loop, the high initial score of 863 points would drop by as much as 25 percent over time. The Core i7-1065G7 in the Dell XPS 13 7390 2-in-1 would behave almost identically.” That quote comes straight out of the more flattering NotebookCheck article. All in all, Tiger Lake is Ice Lake with some process efficiency improvements and a much-needed graphics upgrade. Yet even games are a loss. Here again NotebookCheck doesn’t hold back: “Frame rates when running Witcher 3, DOTA 2, or GTA V are all noticeably slower than the AMD-powered Yoga Slim 7 despite what our 3DMark results above suggest.” I would bet anything they likely optimized their drivers to the benchmark but not to actual real-world gaming performance. Tiger Lake is not at all the slam dunk checkmate win over Renoir-based Ryzen 4700 U that Intel’s marketing department made it out to be. If anything, at best, it’s a stalemate, and if you need multicore performance, it’s an out-and-out complete loss. And with Zen 3 laptops just half a year out, I would strongly discourage anyone from buying Intel 10nm ++ Tiger Lake. Unless you have a specific feature need that is only found on an Intel model of laptop, buy Zen 2/Renoir for right now or simply wait for Zen 3/Cezanne next year. =================================================== If you chose to blindly believe what Intel says at this stage of the game, then you can easily lose all credibility as a reviewer. Don't buy Intel 4 core AI trickery based Tiger Lake 10nm ++. Buy a solid real 8 core AMD Renoir 4700 U instead. Get the better performance and save all that "wasted on Intel tilted AI lies" money as well. Buy you a well tested overall winner laptop, such as the current PC Magazine Editor's Choice. You are the greater part of 6 months away from Intel being able to actually build something actually worthwhile out of Tiger Lake -- and by that time that actually happens an entire new generation of AMD laptop processors will wash it all away like ocean waves hitting a kid's sand castle. And please note that chiplet based AM5 socket 5nm 15 layer AMD processors are coming for real inside a year and a half and will be light enough and low watt power draw enough to fit just fine into the laptop space while swinging 12 -16 cores and a real AI co-processor and tons & tons of on-chip memory. AMD 5nm chiplet based construction will mean a lower cost compared to the current AMD APU construction methods, plus large supplies of chipsets could be readily assembled from chiplet inventories to meet whatever the demand happens to be this week. APU chipsets are really not as flexible as chiplet designs, and APUs are more prone to demand gaps and small shortages type issues. Plus any flaw in an APU means a scrap chipset, and this cost has to be added in to the price calculations as well. SHORT TERM AMD NEEDS Right now AMD needs to get their Renoir 4700U out at their laptop builders with a stock fully supported higher current draw option that is user software selectable. Such machines will need better cooling to support this particular trick, something like the ROG gaming machine and the XMG machine uses. Better cooling and a watt draw set up to 31 to 35 watts allows the ROG and the XMG to kick all potential Tiger Lake arses, and that is a fine place for AMD to be right now. Give the user the user selectable choice of 15 watt level performance, or the choice to speed it all up into ass kicking mode and trash the competition accordingly ..... =================================================== OUT IN THE FUTURE Intel 6nm purchased from TSMC will wind up being Intel's next real chance at doing something to really get back into the game. 3nm AMD timing will be close on this one time-wise, so any Intel 6nm advantage is likely to be only temporary. Intel 10nm ++ is not really competitive if you back out all the blatant Intel lies and cheating that is currently going on ...... Simply planning to skip Intel 10nm completely might be a good idea, as it has been much delayed by years now and Intel 10nm is going to be "overcome by events" very shortly. Intel's poor choices in Intel's cheating and lying PR are currently hurting Intel's image badly, and this image damage really does not disappear as quickly as Intel would like to think. http://https://cdn.mos.cms.futurecdn.net/tkAiCENfeWLkJCY7ZRMrRa-650-80.jpg.we |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/02/20 at 00:56:39 I must reality check myself. Intel is back in the game NOW with Tiger Lake, based upon AI cheats and false "innovations" that Intel's many commercial builders have apparently fully bought into and have incorporated just so they can build some new products to sell for this Christmas. Tiger Lake is going to be the new laptop baseline for Intel and for the laptop industry in general. I've got to get over this "fake testing" roll out, this massive misleading of the purchasing public. It is the new normal. ...... for Intel to lie and mislead people, that is ...... =================================================== Intel does honestly say that Tiger Lake is better than any previous Intel 14nm was and this is true. 10nm ++ Tiger Lake is not any better than AMD 7nm Renoir, it is in fact inferior. Sitll, AMD has to bring forward a MORE COMPELLING (totally crushing) Tiger Lake beater 7nm Renoir version if AMD wishes to continue to take market share away from Intel's lying misleading marketing machine. Compelling enough to break through the big lies Intel has already laid down in people's minds, anyway. Tiger Lake is going to systemically try to teach laptop folks that very high short term current draws and sub-par battery life are "normal" and AMD also needs to put out a 4700U product that can be set to run up at these much higher hyper-clocking performance levels too, ASAP. Heavy AI Usage needs to be totally planned for by AMD, as you can see that heavy AI usage is coming on clearly in a very big way as Intel DEPENDS upon it for so very very much starting right now ..... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/02/20 at 08:09:28 https://www.youtube.com/watch?v=6x8SAAk_J4c it is a video, click on it ---- takes some time but it is worth it FIRST EVER XMG SAME-SAME LAPTOP REVIEW SET Everything I have said before rings out clearly in this detailed video review --- the reviewer points out that most Intel units are carefully tuned to suit the Intel Tiger Lake processor and the AMD units clearly didn't get the same level of attention paid to them or are just put together and are not optimized at all. The vendor in question here, XMG, has carefully optimized both identical battery and aluminum chassis heat pipe based units and softwares, separately, to give very best performance across all tests that is possible for each brand. XMG uses the same sort of heat pipe system on both brands and both can run a very similar continuous watts draw which is quite similar to the ROG gaming unit that we mentioned earlier. The AMD unit consistently kicks Tiger Lake unit's arse by 17% to 54% performance-wise depending on the item being tested. AMD does not get as hot nor does it drink nearly as much battery power. As such it completes tasks quicker and gets more done in the same amount of time. NOTE All tests were run off an unplugged fully charged up battery. SOME INTEL TIGER LAKE TESTS DID NOT FINISH THE TEST AT ALL AS TIGER LAKE DRANK THE ENTIRE 46 WATT-HOUR BATTERY COMPLETELY DRY BEFORE FINISHING THE TEST, WHILE THE AMD UNIT ALWAYS FINISHED THE TEST WITH OVER 25% OF THE BATTERY LIFE STILL REMAINING. And the AMD unit got it all done quicker in all cases, too. This is the level of review I would think all reviews and reviewers should be doing ...... but sadly this is not the case. I hope all the bogus lazy slack reviewers are being very well paid by Intel, as they are selling off their reputations as reviewers permanently as we speak. =================================================== Some laptop makers are equally lazy and slack and just did what they have always done when making up the new Tiger Lake laptops, used the same size batteries, etc. etc. They are now learning exactly WHY Intel recommended that the battery in the laptop be a larger 61 watt-hour battery. Most batteries are not that large, BTW, most are 46-54 watt-hour batteries that work great with normal laptops that are set to pull a 15 watt total draw (as used to be the industry standard). Unless they put in the larger 61 watt-hour battery, they can expect that Tiger Lake will suck their battery dry inside two hours of heavy use. (less if something you do kicks off hyper-clocking repeatedly) Plugging the charger in all the time helps some, if you go with Tiger Lake you will have to make this a firm habit. College students who take their Tiger Lake laptops to class with them will struggle to to sit next to a wall socket in the older classrooms that exist on most college campuses. They will likely have to go on line to learn how to turn off the Tiger Lake hyper-clocking feature as that trick will really suck down their standard laptop battery in jig time ...... I think a lot of normal people will be returning Tiger Lake units to Amazon as defective because of this "defect". When they fill out the Amazon return, they will say "won't keep a charge" or "charge only lasts 2-3 hours". |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/02/20 at 11:32:16 https://www.lenovo.com/us/en/d/deals/clearance-sale?sort=sortBy¤tResultsLayoutType=grid&clickid=2ZMSKDQOOxyOWyIwUx0Mo3YlUkiSd93FxT4T1k0&irgwc=1&PID=10078&acid=ww%3Aaffiliate%3Abv0as6 https://www.lenovo.com/us/en/d/deals/doorbusters?sort=sortBy¤tResultsLayoutType=grid&clickid=2ZMSKDQOOxyOWyIwUx0Mo3YlUkiSd9xNxT4T1k0&irgwc=1&PID=10078&acid=ww%3Aaffiliate%3Abv0as6 The Handwriting is up on the Wall Simultaneously to the AMD APUs coming out in force and taking all the Editor's Choice awards, Lenovo has begun a 40-60% off INTEL clearance sale and a separate 30% off general INTEL sale intended to sweep the obsoleted Intel based units out of their warehouse stock shelves worldwide ...... all the moldy old Intel stuff will be gone by New Years Day. I suspect older Intel products everywhere will star in Black Friday ads all over the world. ::) https://www.lenovo.com/us/en/desktops-and-all-in-ones/thinkcentre/m-nano-series/ThinkCentre-M90n-1/p/11ADS0BB00 =================================================== Keeping good balance in mind, you also have to realize that this intentional sell off of the old Intel products also makes more room for Tiger Lake to move into replace them. WARNING: The Intel PR machine is still claiming that Tiger Lake naturally INCREASES battery life, something we all know now is false. Hyper-clocking crashes battery life, certainly not acting to help it in any way at all. Next, the Intel PR machine is claiming that heavy AI usage is "only a small part" of their Tiger Lake implementation, not as the absolute hinge pin of all their bogus "better than AMD" speed claims. Lastly, hyper-clocking (running 58-62 watt power draws for short periods of time) gets ZERO mention from the Intel PR machine or from any of their tame builders in any of their literature or verbal claims. And we now know that, by default, every Tiger Lake chipset system automatically hyper-clocks when given any load that goes up all the way ...... It is very much like the PC laptop sales machine wants to peddle us this stuff willy-nilly. They are trying to force you to buy it by keeping mum about all the downsides and actively promoting all the BS Intel claims that we all know are STONE FALSE. Concerted attempts are also being made by Intel to intentionally derail the momentum of the vastly superior eight core 16 thread AMD Ryzen 4700U during this initial Tiger Lake sell-in period. These efforts have included a widely dispersed rumor that AMD is having chip supply problems. AMD denies this rumor, and so far we see there has been a marked lessening of the B/O periods showing at Amazon but some delays are still showing on some models where the laptop builders simply did not make enough units to fill the abrupt AMD demand. It is very important for Intel to get a lot of Tiger Lake sold right now QUICKLY, before the battery life issues and overheating issues on Tiger Lake initial units become widely known off the initial shipment of real Tiger Lake units to the purchasing public. Tiger Lake units (especially those sold with less than 60 watt hour batteries) will be prone to dying too soon due to the high current draws associated with Tiger Lake's built-in hyper-clocking functions. ================================================== https://www.reddit.com/r/Amd/comments/e7stwb/intels_false_and_deceitful_marketing_will_amd/ Someone else sees these same things from Intel's BS Marketing ...... =================================================== https://www.amazon.com/Acer-Octa-Core-Graphics-Fingerprint-SF314-42-R9YN/dp/B086KKKT15/ref=sr_1_5?dchild=1&keywords=laptop+with+a+AMD+Ryzen+7+4700U+processor&qid=1601765222&s=electronics&sr=1-5 Intel's bogus Ryzen 7 4700U chip shortage rumor dries up and blows away ....... Amazon can direct drop ship and deliver one at your house by Monday Oct 5. That is 2 day delivery, boys and girls on an item Intel says AMD doesn't even have ........ ::) ...... what part of the INTEL fraudulent black bag methodologies do you need me to explain to you again ???? Let's see, you want a good current example of Intel black bag tricks to illustrate how Intel works now-a-days....... that head of Intel development's post on the avocado toast discussion site that gently mentioning the proposed 8 core 16 thread Tiger Lake chipset that was supposed to be coming --- that post was DELETED yesterday, so that wasn't real either, just another Intel BS black bag trick ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/05/20 at 03:06:32 Boyd Phelps, Intel's Corporate Vice President of Client Computing Group -- has completely pulled his sneaky avocado toast post and replaced it with this carefully stated missive. https://medium.com/intel-tech/11th-gen-intel-core-processors-new-architectural-breakthroughs-b552fc77dce0 Boyd Phelps has also apparently added a new disclaimer thing to all of his formal Intel informational posts in an effort to combat some of the "lying BS" charges that have been laid out against Intel lately. Actually, this disclaimer FUNCTIONALLY CONFIRMS all of the charges that it is intended to deflect, that Intel has indeed done these things and fully intends to do them out in the future as well. It also boldly asserts that Intel has an ongoing right to continue to give out bogus rankings and testing results and to provide this false information to their builders to pass on to the public whenever the heck they want to ...... and I do believe the EU regulators will disagree with this dubious Intel position. ::) Numerous individuals have already logged complaints in the EU against Intel for this active misleading false advertising campaign on Tiger Lake. Rumor has it that the EU is squaring off to whomp Intel with some very large fines over this sort of false advertising nonsense --- and mebbe Intel is trying for some "broad coverage butt protection" here based upon this funky disclaimer. Notices & Disclaimers Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. Performance results are based on testing as of dates shown in configurations and may not reflect all publicly available [ch8203]updates. No product or component can be absolutely secure. Your costs and results may vary. Intel technologies may require enabled hardware, software or service activation. © Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others. "Your costs and results may vary. Intel technologies may require enabled hardware, software or service activation." Boyd, are you saying that unless we pay you service money and let you into our stuff at night so you can fiddle with it, that you are now saying it won't work optimally in the long run? Or, are you saying that unless we provide rapid internet access to all the tweeked Intel background libraries, the ones that all your dirty AI speed up tricks depend upon, that the dirty AI tricks won't work right out in the future and the "performance" you promised us to get us to buy your Tiger Lake garbage simply won't be there when we need it? Which is it Boyd? Pick one ...... (or both if both apply) :P Intel really sux, sorry, Intel just sucks brown slimy ditchwater compared to AMD ..... To have to write a DISCLAIMER like this for your news stories means you ADMIT you have been lying out your ass to the public this whole durn time ...... :P :P :P :P :o No product or component can be absolutely secure. :-/ Boyd, folks haven't stumbled over the exact issue that prompted you to add this one to the disclaimers listing, but we are all keeping an eye out for it now since you clued us all about it ...... ;D Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. ;D This one all by itself has been worth a 10 Million Euro EU fine against Intel in decades past, but seeing all the rest of them all lined up in a row and knowing several of these are third strikes against Intel means the EU fines may be 10x bigger this time around. If I were an Intel builder that has been advertising using these known to be false claims, I'd be afraid that I might be found to be co-defender with Intel under current EU rules. I would then have to prove in an EU court that Intel had lied to me too, actively deceiving me into transmitting their lies in order to recover the money that I was fined by the EU, getting it in the end from Intel by a separate lawsuit. Fun, huh? |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/06/20 at 06:26:41 Watch out when people make comparisons to AMD 4000 processors. The PC MAGAZINE EDITOR'S CHOICE machine is the Ryzen 4700U not the older Ryzen 4800U that does not have two threads per core, and that older 4800U is simply slower in all aspects of I/O and memory, etc. etc. etc. Look here to see what all the differences are. https://www.cpu-monkey.com/en/compare_cpu-amd_ryzen_7_4700u-1093-vs-amd_ryzen_7_4800u-1142 I find that a lot of review houses don't even have a modern Ryzen 4700U on hand to test Tiger Lake against --- so they report against some two year old AMD laptop that they do have because they are simply slack and lazy. Yep, on the Intel side they tend to use the Intel Company Owned Tiger Lake Sample Laptop, or else they just copy info from their competition's review or from the Tiger Lake Press Kit itself. This whole thing is a hairball, with Intel provided tilted misinformation resting down at the bottom of it. =================================================== https://liliputing.com/2020/10/11th-gen-intel-rocket-lake-chips-coming-to-desktops-in-q1-2021.html Intel's queasy marketing methods are continuing, as they are prepping a set of desktop chipsets to follow in this same very dubious marketking pattern. Issue with desktops is that there is NO CLOSE PARITY IN DESKTOP CHIPSETS and AMD has already crushed Intel's best and everybody knows it. Plus, trying to tweek desktops with AI enough to make up lost ground will be Tough To Do as you are trying to recover a 2x advantage in a lot of areas. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/07/20 at 19:44:03 https://liliputing.com/wp-content/uploads/2020/10/arm-makalu-768x368.jpgl http://https://liliputing.com/wp-content/uploads/2020/10/arm-makalu-768x368.jpg ARM is planning to roll up to 64 bit on all cores by the time 5nm is rolled out completely inside of 2 years from now. This is not "new" news and any vendor that gets left stuck on 32 bit code by then has been on vacation for 5 years already. And yes, ARM is once again waving a "massive simplification" in the "Big / LITTLE" coding of such unified 64 bit critters, and ARM is also saying there will be a 30% "coding performance upper" to be had overall on top of the lithography bump that comes from 5nm ...... Raw code size will go down too, and AI will become easier to implement. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/08/20 at 11:43:35 https://liliputing.com/2020/10/amd-zen-3-desktop-chips-coming-nov-5-for-299-and-up.html Intel, I warned you not to go piss off Lisa Su --- You won't like her when she's mad ....... "....... I'll rip off your last crown and take your head clean off with it. I'll stuff it in a box along with a fresh steaming example of Intel Marketing dumped all over it ....." http://https://cdn.mos.cms.futurecdn.net/FnUhepxnuLSod6zUdNDtiF-1200-80.jpg Intel’s 11th-gen “Rocket Lake S” desktop processors may be coming early next year. But rival AMD’s newest desktop chips are coming next month on Nov. 5th. AMD has unveiled four new Ryzen 5000 series desktop processors with prices ranging from $299 to $799 and the promise of major improvements in single-core performance, multi-threaded performance, and energy efficiency. The new chips are also the first CPUs to feature the company’s new Zen 3 CPU core architecture. Here’s a preview of the launch lineup: Cores / Threads Base / Boost freq Cache TDP Price Ryzen 5 5600X 6 / 12 3.7 GHz / 4.6 GHz 35MB 65W $299 Ryzen 7 5800X 8 / 16 3.8 GHz / 4.7 GHz 36MB 105W $449 Ryzen 9 5900X 12 / 24 3.7 GHz / 4.8 GHz 70MB 105W $549 Ryzen 9 5950X 16 / 32 Up to 4.9 GHz Boost 72MB 105W $799 The new chips are 7nm processors with a new core layout for better communication between cores, reduced memory latency for gaming, and a 19-percent boost in instructions per clock as well as up to a 24-percent improvement in energy efficiency for up to a 24-percent boost in performance per watt when compared with the Ryzen 3000 line of desktop processors. http://https://liliputing.com/wp-content/uploads/2020/10/zen-3_04-768x527.jpg Up until recently, AMD’s desktop chips have held the lead in multi-threaded CPU performance while Intel processors had an edge in single-core performance. But AMD says it’s improved its single-core performance enough that its Ryzen 9 5900X processor can get a single-threaded score of 631 in Cinebench, while the Ryzen 9 5950X can snag a score of 640. That would indeed put them at the top of the charts. But for now we’ve only got AMD’s word to go on. We won’t have long to wait for independent testing though, with the chips set to hit the streets in less than a month. A Ryzen core complex used to be made up of sets or groups of 4 cores per set. Now it is laid out in tighter groups or sets of 8 cores and the new 8 core set has inherent 19% better IPC and 24% better energy efficiency. This means roughly around 30% higher clock rates should be available before serious overheating occurs. Better coolers could stretch this even further ...... The center core swings more memory layers, plans for a separate AI layer and FAR BETTER ON-CHIP GRAPHICS (if provided -- the largest core count units will lack a on-board GPU since they are intended for gamers who always want separate very high performance video cards). |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/08/20 at 21:34:13 Confirmation and "deep dives" into real detail levels ........ https://www.anandtech.com/show/16148/amd-ryzen-5000-and-zen-3-on-nov-5th-19-ipc-claims-best-gaming-cpu https://www.theverge.com/2020/10/8/21506447/amd-zen-3-ryzen-5000-processors-gaming-cpu-price-release-date https://www.cnet.com/news/amd-unveils-ryzen-9-5950x-gaming-pc-cpu-with-16-cores-zen-3/ Read the Anandtech article, it is the best of the lot. More info will be released next month when the chipsets ship, mostly having to do with clock speeds and thermal setups I would think. http://https://images.anandtech.com/doci/16148/1%20Slide%207_575px.jpg |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/09/20 at 09:44:26 Busy day for AMD ...... https://www.edn.com/why-amd-is-acquiring-the-fpga-powerhouse-xilinx/ http://https://www.edn.com/wp-content/uploads/media-1174085-virtex-7-h580t-02a.jpg Why XILINX?? Re-programmable fast automotive chipsets and AI ACCELERATORS. Xilinx is the best in this field by far. And, please remember so far this tech has not blended in with PC tech completely (yet). AI has just started to show up in testing fudged by Intel and the only applications that uses AI routinely is graphics (example, while resizing images). I see Xilinx tech as a well integrated separate layer inside each AMD multi-layer chipset, eventually, blended in seamlessly as a common L3 cached CPU/GPU/AI Accelerator. Re-programmable over the internet, no less. On-going net optimized accelerator code sounds really really neat to me. I see progress happening ...... lots and lots of it at the 5nm 14 layer lithography level. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/09/20 at 23:59:07 So far I have watched response videos from 3 good major review houses for the upcoming Ryzen 5000. In all cases, they say the same thing. "AMD generally does what they say they will do, legitimately". They all say the same thing "AMD said to wait for the Ryzen 5000 chips to land at retail November 5th, then pick one from the flow, test them for real and report what you see." This is very very different from the current Intel Marketing BS methodologies which depend on tweeked hyper-clocked Intel owned test units, hidden driver libraries, intentional unit delays and lots & lots of AI trickery. Oh, let's not forget the nice long Intel DISCLAIMER LISTING ...... :P All the review houses have both old and new AMD motherboards on hand (as this new chip supposedly plugs in to either sort and runs, but runs best on the most modern boards). We are being told now that the upgradable organization ideas of the current AMD motherboards and sockets and chiplet groupings will carry over into the new AM5 socket for 5nm lithography that is coming next year which will have over 5 mhz speeds natural to it, combined with the wide data paths, the current large IPC uplifts and all the latest AMD neat stuff will be carrying over into 5nm lithography. Or, it may be more accurate to say the newest organizational ideas started in the AM5 socket 5nm lithography product and are backflushing into the current Ryzen 5000 7nm product at this time. We are told by each of the pundits that Intel is likely toast in both content creation and in gaming now and Intel will likely remain toast going forward, and Intel will not recover any crowns, ever, no matter what shady tricks they pull with bogus AI testing tricks and other Intel dirty black bag tricks. The gap for Intel to make up is currently around 30-40% and that is literally too much of a performance gap to lie your way through. :o Pricing has flipped, Intel is now less desirable and commands lesser price points compared to AMD. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/10/20 at 22:39:46 Folks are fact checking chip sales positions using Amazon as the data source. Price Options [#] is Amazon Best-Seller Position # AMD Pt # Price [1] Ryzen 5 3600 $205 [2] Ryzen 7 3700XT $295 [3] Ryzen 5 3600X $209 [4] Ryzen 5 2600 $149 [5] Ryzen 9 3900X $430 [6] Ryzen 7 3800XT $340 [7] Ryzen 5 3600XT $230 [8] Ryzen 7 2700X $218 [9] Ryzen 3 3200G $100 We shall see how this all shifts out when the newer chips land and the first reviews come out. And no, you are not completely nuts, there is no Intel on the current Amazon customer top 10 wish list. ========= A Big Big difference exists between individual chipset sales and entire built up unit sales. Intel in fact is still selling a whole lot of chipsets to brand name unit builders as they will keep building the existing model line filled with Intel products until model change time comes around. Major builders seem to need Intel's slower pace of improvements as they do a model change per year and a cosmetic change or two inside of a year and that is all the change they are structured to handle. Next, the computer industry as a whole cannot flush out all these new models fast enough to keep up with AMD's change rate. Inventory is accustomed to lasting 1-2 years in the sales channels before going obsolete. The phone industry is closer to what AMD requires, rapid multiple unit types designed and built very quickly and flushed just as quickly. AMD is currently overstressing the complete computer box builder's rapid response capacity right now. It is good that AMD keeps a consistent socket for multiple years as this gives builders some chance to keep the same computer guts for 2 years running and just plug in the new AMD chipset and BIOS for each wave of new. AMD improves too rapidly. AMD has complete model build needs that are too erratic and too fast acting, especially if they hit gold on a unit and the box builder hasn't got the capacity to even start to meet that unit demand before the next change wave rolls up. The computer industry is building lots of finished boxes that folks simply don't want right now as they slavishly follow the old Intel build patterns from years past. ================================================== Woot is going to have to take a WHOLE LOT of virgin boxed units to sell this Christmas over this huge "build lag" issue ........ Look to see Amazon using Woot and "Warehouse Specials" a lot this Christmas. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/15/20 at 09:27:38 https://liliputing.com/2020/10/chromebook-lifespans-creeping-upward-some-models-get-9-years-of-os-updates.html http://https://liliputing.com/wp-content/uploads/2020/10/c13_01-768x570.jpg Chromebook lifespans creeping upward These Chromebooks are some of the first to feature AMD Ryzen processors, and both models were launched in September of 2020, which means they’ll be supported for more than 8.5 years. Chromebooks cost a good bit less than Windows machines (by and large). A good Chromebook can run ChromeOS, Linux, Android and (using CrossOver Chromebook) it can run MS Office as well. Also note that all Chromebooks can run Libre Office as is, and that means they can open any generation of MS Word files, change it and then put it back the way it found it. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/16/20 at 09:11:47 https://www.androidauthority.com/graphene-batteries-explained-1070096/ https://www.youtube.com/watch?v=dnE1nO6o-do click and watch --- graphene going commercial is a BIG DEAL for phones, laptops and EV cars ...... Graphene offers higher electrical conductivity than lithium-ion batteries. This allows for faster-charging cells that are able to deliver very high currents as well. This is particularly useful for car batteries, for example, or fast device-to-device charging. High heat conductance also means that batteries run cooler, prolonging their lifespan even in cramped cases like a smartphone. EDITOR'S PICK Solid-state battery — the successor to Li-ion — takes another step closer to our smartphones Graphene batteries are also lighter and slimmer than today’s lithium-ion cells. This means smaller, thinner devices or larger capacities without requiring extra room. Not only that, but graphene allows for much higher capacities. Lithium-ion stores up to 180Wh of energy per kilogram while graphene can store up to 1,000Wh per kilogram. Finally, graphene is safer. While lithium-ion batteries have a very good safety record, there have been a few major incidents involving faulty products. Overheating, overcharging, and puncturing can cause runaway chemical imbalances in li-ion batteries that result in fire. Graphene is much more stable, flexible, and stronger, and is more resilient to such issues. You don’t have to have one or the other though. Li-ion batteries can use graphene to enhance cathode conductor performance. These are known as graphene-metal oxide hybrids. Hybrid batteries result in lower weight, faster charge times, greater storage capacity, and a longer lifespan than today’s batteries. The first consumer-grade graphene batteries are likely to be hybrids. What graphene batteries mean for smartphones Future smartphones packing graphene power cells would exhibit the benefits outlined above. Handsets would charge even faster (check out the results when we tested a graphene powerbank in the video at the top of this article), battery life would easily last a day or two, if not longer, and devices could be thinner and lighter. The move to graphene could offer 60% or more capacity compared to the same sized lithium-ion battery. Combined with better heat dissipation, cooler batteries will extend device lifespans too. You won’t need to pay for expensive battery replacements after a couple of years to keep your old devices performing in top condition. https://www.youtube.com/watch?v=uIMegpibt1M here is your obligatory Gary Explains Graphene YouTube video |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/16/20 at 17:43:07 Apple's A14 chipset at 5nm is now shipping, in an Ipad no less. This means early next year the 5nm phones come out, and later next year AMD 5nm chiplets will be built im mass, something that will continue ongoing for YEARS and YEARS until 3nm is ready. And here is our buddy Gary, to explain what little is known at this time. https://www.youtube.com/watch?v=F_IsBD6E3O4 |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/16/20 at 22:58:53 https://liliputing.com/2020/10/0renoir-mobile-tuning-app-tweaks-ryzen-4000-laptop-settings-unofficially.html There are no official tools for overclocking AMD’s Ryzen 4000 series laptop processors. But an independent developer has come up with the next best thing: an unofficial Renoir Mobile Tuning app that gives you control over the power and temperature limits of AMD’s Ryzen 4000 chips for notebooks. http://https://liliputing.com/wp-content/uploads/2020/10/ryzen-mobile-tuning-768x517.jpg “Renoir” is the code-name for the AMD Ryzen 4700U and Ryzen 4800U series processors that launched earlier this year and the “Tuning” part comes in because the application allows you to monitor and adjust settings that can help extend your laptop’s battery life or boost performance. Among other things, the app can provide detailed information about your computer’s clock speed, power consumption, thermals, and power limits. You can also adjust those limits or change how long the processor can run at boost speeds, although if you over do it your laptop could overheat and throttle the speeds. AMD has been slow to react to Intel's BS manipulation of Tiger Lake Test Units, so now the enthusiast community has filled in that need with a hacker tool. AMD needs to provide a built-in solution with reasonable limits, as letting enthusiasts "toast up" a bunch of Renoir laptops requiring warranty replacement is not in AMD's best interest. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/17/20 at 13:37:02 https://www.anandtech.com/show/16084/intel-tiger-lake-review-deep-dive-core-11th-gen/19 I mentioned that Anandtech had promised a THROUGH review of all Tiger Lake functions. Over the weeks since then new web pages of results have been released off this same original review, totalling 19 web pages of Anandtech testing results. Exhaustive testing was done on each of the BS Intel Marketing claims and each one was dissected separately and completely. (look for the drop down list of test pages to get to this information) After literally a month of detailed analysis work, Anandtech has reached a conclusion. by Dr. Ian Cutress & Andrei Frumusanu, Anandtech Magazine condensed Conclusion: Is Intel Smothering AMD in Sardine Oil? The new Tiger Lake still falls down against the competition when we start discussing raw throughput tests. Intel was keen to promote professional workflows with Tiger Lake, or gaming workflows such as streaming, particularly when measured at 28 W or higher boost levels rather than at 15 W industry standard as is used by AMD. AND WE REPEAT: The new Tiger Lake still falls down against the competition when we start discussing raw throughput tests. Intel was keen to promote professional workflows with Tiger Lake, or gaming workflows such as streaming, but ONLY when Tiger Lake was run at 28 W rather than at the 15 W industry standard as was used consistently by AMD. Do you get the point? ALL of the "critical" Intel testing ??? wins ??? were done at an intentional 13 watt + over normal voltage & current draw levels ....... (and you could just as easily say "at double the industry standard amperage draw" and be within 2 amps of totally correct) Despite this we can easily see that the 15 W Renoir options with eight cores can blow past Tiger Lake in a like-for-like scenario in our rendering tests and our scalable workloads. The only times Intel scores a win is due to accelerator support (AVX-512, DP4a, DL Boost). On top of that, Renoir laptops in the market are likely to be in a cheaper price bracket than what Intel seems to be targeting. Only if Intel can convince software developers across the globe to jump on board with using its accelerators in all existing software only then will the customers get full benefit from Intel’s tweeked metrics. The holy grail of this impossible trick may be when it comes to Intel's OneAPI, enabling programmers to target different aspects of Intel’s eco-system under the same toolset. However OneAPI is only just entering v1.0, and any software project base building like that requires a few years to get off the ground. And right now OneAPI only affects Intel's own software packages. We didn’t have too much time to go into the performance of the new Xe-LP graphics, although it was clear to see that the 28 W mode does get a good performance lift over the 15 W mode, perhaps indicating that DG1 (the discrete graphics coming later) is worth looking out for. Against AMD’s best 15 W mobile processor and integrated graphics, our results perhaps at the lower resolutions were skewed towards AMD, but the higher resolutions were mostly wins for Intel - it seemed to vary a lot depending on the game engine. by Dr. Ian Cutress & Andrei Frumusanu, Anandtech Magazine As a concept, Tiger Lake’s marketing frustrates me. Not offering apples-to-apples data points and claiming that TDP isn’t worth defining as a singular point is demonstrating the lengths that Intel believes it has to go to in order to redefine its market and to obfuscate all direct AMD comparisons. =================================================== There was a time and a place where Intel apparently felt the need to share everything, as much as possible, with us as the computer press. During this period Intel actually led us in a group think tank discussion to sculpt the story of where we envisaged the market was going, and Intel OEMs/builders/customers were on hand to add their comments about the viewpoints of the customer base from their perspective. It then let us, as the press, filter back with comments, critiques, and suggestions. This new "cooperative" twist from Intel’s client division, one that’s actually been progressing along this quagmire path for over a month now, will only serve to confuse its passionate customer base, its enthusiasts, and perhaps even the financial analysts. It is clear that by this "request for help" Intel has attempted to co-opt the major review houses as a group to HELP INTEL CRAFT THEIR AI BASED BS "PROCESSOR PERFORMANCE STORY" in a way that the major review houses would then support. This fraudulent conspiratorial action, by itself, is grounds for a BRAND NEW very large EU fine for Intel, in my opinion. IMHO, Intel was in essence attempting to build a conspiracy to commit fraud ....... It is clear that not all of the review houses were receptive to Intel's little scheme. So, Anandtech for one wasn't drinking any of the Intel kool-aid ....... RECOMMENDATION: Skip the Intel 10nm generation COMPLETELY, it is based on a bogus foundation of rigged over wattage hyper-clocking lies, marketing exaggerations and outright AI trickery. Tiger Lake acts much like a FRAUD in motion ...... ..... and Intel will eventually figure out that pissing off the entire computer reviewer community by asking them to lie isn't the best move they have ever made ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/19/20 at 12:46:42 https://liliputing.com/2020/10/is-this-the-first-laptop-with-intel-dg1-discrete-graphics.html Smoke and Mirrors Alert !!!! The next wave of Intel "progress" is breaking as we speak. http://https://liliputing.com/wp-content/uploads/2020/10/tp470ez-768x575.jpg According to the project page for the Asus VivoBook Flip 14 TP470EZ, the system features the “First Intel Discrete Graphics.” That vague language is similar to the wording we normally see when a company announces a laptop that will feature an as-yet-unannounced CPU or GPU. But while we usually see things like “latest Intel processor” or “next-gen AMD CPU,” this time Asus is using “first” as a descriptor. Right now it is COMPLETELY UNCLEAR as to what Intel is attempting to advertise. This falls in line with recent Intel Marketing efforts, leaving things totally in the dark and making big vague claims about "first' and "next-gen" and "latest". This new thing may range from a Tiger Lake laptop implementation to a 10nm discrete GPU tagged on to a existing Intel mainstream 14nm laptop chipset. In either case it deserves a Smoke and Mirrors Alert, loud alarm horn and red flashing lights ...... It is Intel, so who really knows what the heck it really is ???? As the buyer of this thing you know two items going in, first that there is a REASON Intel won't tell you what it really is up front, and second that you are not going to like the mundane reality that lies behind all the splash and glitz ....... ================================================== https://wccftech.com/asus-vivobook-flip-flip-14-unveiled-first-intels-dg1-discrete-iris-xe-max-graphics-tiger-lake-10nm-cpu/ http://https://cdn.wccftech.com/wp-content/uploads/2020/10/ASUS-VivoBook-Flip-14_Intel-Tiger-Lake_Intel-DG1-Discrete-Graphics-GPU_7-740x482.png The GPU specifically is referred to as the Intel Iris Xe Max as identified over at ASUS's website by Momomo_US. Benchmarks of the Intel Iris Xe Max graphics chip appeared a few days ago at the Compubench database (via TUM_APISAK) which shows the Max chip slightly ahead of the standard Iris Xe GPU featured on the Tiger Lake CPUs as integrated graphics. The chip also leaked in the SiS ofware Sandra database along with a range of benchmark numbers. ....... whups, a whole bunch of yellow splash and glitz over some "slightly ahead of Tiger Lake CPUs", huh? You know, the same Tiger Lake CPUs that are slightly behind AMD APU GPUs at the moment ...... Sounds a bit "equivalent" doesn't it ??? :P All of Intel 10nm shares a common theme ---- it was delayed 5 years and three generations while Intel struggled mightily to get it to work at all ---- and that long delay left a mark on all Intel 10nm products. They are all "a day late and a dollar short" today in 2020-2021 when they finally get introduced. Even 2 years ago they would have been great stuff, but not today ...... What is the real current great stuff is a six core TSMC 5nm processor that is shipping today (in Apple iPads) ....... Look at the 5nm tidal waves coming in the next six months, they will wash all relevance away from the Intel's little 10nm sandcastle (what little relevance it actually has, anyway). |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/21/20 at 01:44:32 https://www.xda-developers.com/us-government-antitrust-charges-google/ Attorney General William Barr has issues with Monopolies, and he quite competently has determined that Google has become one (yet again). http://https://www.xda-developers.com/files/2020/10/mobile-search-engine-us-market-share-september-2020.jpg Google has already went through this recently with the EU, twice, and the implemented EU remedies will provide Google's main defense in this USA Monopoly trial. The harsh fact is that once you give everybody a free choice, they will still pick Google 94% of the time because it works so much better than the alternatives. No joke, if you don't provide a path to load Google, users will find a way to SIDE LOAD it because they bloody well want it ....... I just went through all the steps of side loading Google on my Amazon Fire devices, so I know what of I speak. This anti-trust stuff always winds up always being a money play with the EU anti-trust actions, Google tunes a few things, pays a few 10's of Millions in fines and removes some restrictions intended to keep fragmentation at bay, then later on gets politely requested by the EU to put the same restrictions back in place later when Gooble services all starts to unravel in the EU due to the EU fiddling with things they did not really understand. Google provides the checks and balances on Microsoft and Apple, and same thing can be said backwards for Google. There is no real enforced monopoly here, simply strong personal preferences showing itself all over the place again. Barr may decide to bust Google up into separate companies ---- except it already is structured as separate companies and dicing it up further into divisions=companies is sorta meaningless to a communications company like Google. He can do that though. He can force them to compete with each other too, like they already do for the available resources inside Google. Google as a product is still better, and is still prefered by most people. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/21/20 at 02:42:39 https://www.cnx-software.com/2020/10/19/intel-celeron-and-pentium-tiger-lake-processors/ Intel Celeron 6305 and Pentium Gold 7505 Tiger Lake Processors Launched Intel launched various Core i3/i5/i7 Tiger Lake processors in early September, shortly followed by more Core i3/i5/i7 UP3 Tiger Lake Embedded SoCs, so I was a little confused when Congatec launched a Tiger Lake COM Express module with of choice of processors including the usual Core processor plus an unnamed Celeron Tiger Lake processor because none were listed in Intel Ark. But a few weeks have passed, and there are now three new Celeron and Pentium Tiger Lake processors in the list namely: Pentium Gold 7505, Celeron 6305, and Celeron 6305E with the later optimized for embedded & IoT applications with a wider temperature range from 0 to 100°C. http://https://www.cnx-software.com/wp-content/uploads/2020/10/Celeron-Pentium-Tiger-Lake-768x512.jpg.we Intel has gotten some existing designs to run on its 10nm lithography lines. Not a lot of benefit to doing this, but here it is in all its glory. In some cases the original 14nm product actually works better, so Intel will still sell that 14nm product as part of the "10nm family of products". It isn't really 10nm but they will sell it as such anyway ....... Lastly, Intel plans to let these things get on up to 100°C (that's 212oF) in use in a thin and light laptop so they will "perform better". You do realize that 100°C means as hot as boiling water blowing on your legs and your lap ?????? OUCH !!!! Talk about sucking down a laptop battery in a hurry, trying to flash boil your miscellaneous man bits will certainly do that for you. =================================================== I find this funny ...... Intel says they are coming out with an 8 core Tiger Lake late next year sometimes or another to "crush AMD's APU products". The post has some vague big promises attached to it, but it gets deleted after showing for only two days on an avocado toast discussion web site. So, OK, you went and pissed Lisa Su off and she is going She-Hulk all over your arse. You were warned, remember? Hurts, don't it? Difference between Intel's vague plans and Lisa Su super compacting your spine for you is how close to arms length Lisa Su can get in a hurry ....... (ouch) ...... like month after next? CES is January 11th which is easy for me to remember (its my birthday). http://https://cdn.mos.cms.futurecdn.net/FnUhepxnuLSod6zUdNDtiF-1200-80.jpg https://videocardz.com/newz/amd-ryzen-5000u-cezanne-lucienne-mobile-series-specifications-leaked The ‘flagship’ Ryzen 7 5800U SKU is said to feature 8 cores and 16 threads. This processor will feature 8 GPU Compute Units clocked at 2.0 GHz. The CPU will have 16MB of L3 cache, a double of what Ryzen 4800U and Ryzen 5700U have to offer (both 8C/16T SKUs based on Zen2). Interestingly, all SKUs in Ryzen 5000U series are to feature SMT, which was not the case for Renoir. The predecessor had three parts with SMT disabled. According to the leak, AMD is set to launch Cezanne/Lucienne-based AMD Ryzen 5000U series in 2021. We might be hearing more from AMD around ‘Virtual’ CES 2021, possibly with the first laptops deploying in the first quarter. |
|
Title: Re: AMD & others --- post AMD dominance in 2020 Post by Eegore on 10/21/20 at 20:20:17 Why the female Hulk images? |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/21/20 at 23:33:51 Super threatening female is reaching out to rip your last gaming performance crown right off your head (hair & head included) ....... those are bits and pieces of your old glory throne that she grabbed on to and smashed into rubble are still floating up in the air behind her --- fragments of your old Intel glory throne that you left behind as you frantically ran and dodged trying to stay out of her grip. Intel understands all of it just fine. Intel would be happy if she just gently separated that last gaming crown and left their head and their scalp pretty much intact where it is. Likely she won't though, she's really really pissed right now. I believe she said something about laying down some freshly laid steaming Intel advertising right down on top of Intel's bare bleeding skull before sealing up the box ...... (really really pissed off women, you know how they get, right?) You realize that for the laptops to be actually be ready to ship by then, TSMC has to have already run them and TSMC has to have already shipped the chiplets and the larger APU units on to AMD for testing and to give some time for AMD to assemble them into fully tested socketed chipsets. Then AMD has to have time to send them the finished tested chipsets on to the build vendors for unit assembly and testing and warehousing ...... yep, all this is in motion right now. :P ...... poor Intel, losing this last crown is really gonna hurt a lot ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/24/20 at 01:31:37 Going over to the flip side of things, the Intel side. Intel has been very successful with their BS Marketing campaign, their pet vendors have flooded the world with AI based and over voltage & wattage hyper-clocking based claims of equivalency (and even outright claims of Tiger Lake superiority which are even more dubious). Intel wants everybody to consider their AI tricks as part and parcel of processor performance, instead of a bias implemented set of software tricks (i.e. intentional cheating). For every honest review put out that focuses on actual processor performance, it quickly gets shouted down and muddled by a host of "other" reviews. Shady reviews and small shady review houses seem to abound right now ....... Customers now seem to depend more on Amazon Customer Rankings than real computer magazine reviews right now since they know they cannot trust traditional computer reviewers any more, but Amazon Customer Rankings have always responded more to the very last sets of rankings received than to the historical mass of rankings on a given processor so that particular trick isn't working right either. Some people now seem to make a living by putting bogus reviews up on Amazon now-a-days, it has now gotten so bad that Amazon now lists whether or not the reviewer had ever even bought the product in question (right up top in the review header). Based upon a thick greasy churning sea of lies, Tiger Lake is doing quite well. ONCE AGAIN, to regain the position lost by intentional BS Intel PR, AMD is going to have to put out a whole new series of clearly dominating processors and get laptop builders to build them into units in large quantities very quickly to keep on taking market share away from the Intel PR machine. This will be hard, as Intel is working on the suppliers themselves again, trying to get them to only use Intel chipsets (illegally again). This sort of exclusivity deals were made illegal in the EU a year ago, but are coming back again because Intel feels them to be worth the risk to pursue them as no one in the EU is really enforcing the last set of decrees and actions against Intel. Covid 19 is taking up all their attentions ..... Intel has become emboldened by their new PR tactics, the BIG Artificial Intelligence LIE is working well for Intel right now and the BIG Artificial Intelligence LIE really is intel's only path forward at this time. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/27/20 at 15:32:02 https://www.amd.com/en/corporate/xilinx-acquisition Well boys, it ain't jest a rumor any longer ....... Expect great things from Lisa Su's Hulked Up new company. 13,000 Best-In-Class Engineers $35 billion in stock exchanged (no $$ outlay needed) All stock transaction New total larger enterprise value Total Addressable Market is is $110 billion in size All Stock Merger Style Transaction minimizes debt and cash outlay Ownership Post-Close AMD Shareholders: 74% Xilinx Shareholders: 26% Leadership Dr. Lisa Su, Chief Executive Officer Devinder Kumar, Chief Financial Officer Victor Peng, AMD President Expected Close 2021 Calendar Year End http://https://www.edn.com/wp-content/uploads/media-1174085-virtex-7-h580t-02a.jpg ================================================== Tiger Lake has been successful in the market place so far based upon AI tricks and hyper clocking a four core not really very great fairly old design of processor, using a slightly older type of graphics core that is being pushed up into overheating to get some higher performance out of it. Intel has "main forced" some Altera tech into a successful first integrated product, but Altera tech is not the nearly the market leader in AI accelerators for Compute nor for Graphics nor for Mainframe. Altera is an "also ran company" in its industry, and is not noted for innovation or for great things in the past. All those crowns (and many others) belong to Xilinx, the company that innovated and created those markets in the first place. What I see happening here between AMD and Xilinx is a friendly corporate merger between two very innovative companies that complement each other. I think Lisa Su's leadership brings the best out in her engineers and they learn how to be ground breakers and innovators under her tutelage. Both sides of the merger are expecting synergy and greatly broadened market bases In contrast, what Intel has done is to normalize lying to the public and making secretive uses of background services to give only their brand of processor something that looks like performance, but is just very biased hyper clocked AI benchmark trickery instead of real innovation. I don't think Intel can continue this long term as their secrets are out now and folks realize that poor battery life and 100-110 oC CPU processors and even hotter GPU graphics processors ARE NOT what people want to be sitting on their laps jest a sizzling & a cooking on their crods. :o |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/29/20 at 11:53:00 https://liliputing.com/2020/10/intel-rocket-lake-s-chips-will-bring-intel-xe-integrated-graphics-to-desktops-in-2021.html Intel Rocket Lake-S chips will bring Intel Xe integrated graphics to desktops in 2021 Despite the fact that Intel has moved to 10nm for its laptop processors, Rocket Lake-S chips will be manufactured using a 14nm process. But Intel says Rocket Lake’s “Cypress Cove” architecture combines features found in its recent 10nm chips, including Intel Xe graphics (which debuted in 11th-gen Tiger Lake processors) and technology from 10th-gen Ice Lake’s Willow Cove CPU cores that have been backported to 14nm. You can also expect much higher power consumption from these desktop chips – while Ice Lake and Tiger Lake processors tend to top out at 25 to 28 watts, Rocket Lake-S chips will likely have a TDP of up to 125 watts. According to Intel, the new chips will offer up to 50-percent better graphics performance than an equivalent 10th-gen Comet Lake-S processor, and “double digit” performance gains in instructions per clock (IPC). Other features include: Up to 8 CPU cores and 16 threads Up to 20 CPU PCIe 4.0 lanes New overclocking features Support for DDR4-3200 memory Support for up to three 4K/60Hz or two 5K/60Hz displays Intel is also promising improved AI performance thanks to new Intel Deep Learning Boost technology. Translation: more Intel 14nm lithography yet again, but this time using all the dirty tricks piloted by Tiger Lake including hyper clocking, extensive use of AI and over volting and over amping of all major items during benchmark runs. Expect very high running temperatures as both the hyper clocked CPU and the GPU will run very hot ...... :P But the company did give us a clue about power consumption by explaining that its performance comparisons are what you’d get if you pitted a computer with a 125W Rocket Lake-S 8-core processor against a machine with a 125W Intel Core i9-10900K 10-core Comet Lake processor. If Intel IPC has "double didgeted" in reality, much better throughput will seem to be present with these chipsets (in general compute tasks, anyway). This is not seen as happening so far. However, Intel has figured out how to muddle up all forms of benchmarking now using AI effects, so expect your benchmark results to lead you off into yet more layers of confusion. Intel counts on this confused AI mess now, lying voraciously in an organized manner to make sure confusion takes place and remains in place. In an AI laced environment, Intel will claim 2x to 4x improvements in certain items (mainly graphics effects). HOWEVER, Intel will make these as "general IPC claims" on the box, knowing Joe and Rita Six Pack really don't care all that much for the fine details when comes to computer shopping ...... Do not trust Intel to tell you the truth any longer, as the truth is simply not in Intel's best interest any longer. "Intel Marketing" is Intel's main progress maker now-a-days. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/29/20 at 20:38:56 As AMD ramps up towards providing full AI acceleration layers in their post Xilinx merger chipsets and AMD begins to ring in 4 threads per core and several other massive 5nm multilayer IPC throughput improvements, the industry will have to learn how to separate out these various effects in testing to be able to report on them accurately. When this level of testing knowledge begins, Intel Marketing ends ...... Intel Marketing needs a lot of "confusion" to exist right now, so Intel has provided itself lots of confusion, bunches of it ........ Apple's call for an impartial control body to set benchmark standards and to control benchmark testing sounds better and better and better to me all the time. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/30/20 at 05:43:56 https://www.datacenterknowledge.com/amd/xilinx-stronger-amd-or-without-it http://https://www.edn.com/wp-content/uploads/media-1174085-virtex-7-h580t-02a.jpg Xilinx’s Adaptive Platform “Traditional CPUs are not able to provide the performance scaling required for some of these compute-intensive workloads,” Xilinx senior VP Salil Raje said at this time last year during the company’s XDF conference in San Jose, the company’s home base. “That is why domain-specific accelerators are stepping up to take on some of these larger workloads,” he continued. “The issue is, however, that the workloads continue to change. They depend on algorithms and standards that continue to evolve. That is why adaptable devices, such as FPGAs and ACAP [Adaptive Compute Acceleration Platform] devices, have started to gain momentum as adaptable, domain-specific accelerators. They can be customized for your domain, for your application, and they can be configured on the fly.” ACAP is a name Xilinx has given to what it considers a new class of processor – if that wording rings familiar, it’s because that’s what Nvidia said it would build together with Arm – in which programmable logic, built-in logic (in the form of “digital signal processor” or DSP, engines), Arm Cortex-A72 cores, and programmable networks reside in attached blocks on the same processor. The re-programmable, “adaptable” part are the concept’s selling point. Unlike a traditional ASIC, where an upgrade involves a truck roll, an ACAP would evolve in place. For now, ACAP is being marketed under the umbrella brand Versal. The AMD and Xilinx deal is about the convergence of CPU, GPU, and other accelerators on a system or motherboard, Kurt Marko, principal analyst at Marko Insights, told us. “FPGAs provide the flexibility to implement a multitude of features that can augment x86 cores by offloading and accelerating particular functions.” The Xilinx platform is comprised of well-respected programmable logic boards governed by a system-on-a-chip (SoC) called Zynq. A programmable logic board can be called an “accelerator,” if you have a system that’s geared to run applications that use the exact same software routines repeatedly, such as AI, video transcoding, or encryption. Rather than rely on a CPU to run these routines in DRAM, an FPGA card may already have those routines stored and running in highly available local memory. All of the major Compute makers are pairing off with an AI company. AMD has Xilinx, ARM has Nvidia, Intel has Altera. All of them are looking forward 5 years to when there is really nowhere to go with shrinking the basic lithography in any meaningful fashion. So, in the end you pick the pair that you feel you can trust the most, and then you go forward into the future with it. ::) quote from Harry Potter book #3's talking shrunken head on the Wizard's bus "Hang on tight, it is going to be a bummmpy ride ........" |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/01/20 at 01:07:12 https://www.webmd.com/skin-problems-and-treatments/news/20101004/laptop-risk-toasted-skin-syndrome#1 All of this refers to old laptops with exhaust air that only ever got up to 115o F ...... What do you think 200o F exhaust air will wind up doing to your skin? long term 2nd degree burn blisters Intel Progress, wonderful stuff ain't it? ::) I am looking for Intel's builders to change the air flow and air ejection patterns of their laptops AND ADD SKIN SIDE INSULATION of some sort to handle this 2nd degree skin burn issue ..... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/02/20 at 08:27:18 https://liliputing.com/2020/11/raspberry-pi-400-is-a-pc-in-a-keyboard-for-70-and-up.html Raspberry Pi 400 is a PC-in-a-keyboard for $70 and up Raspberry Pi has been selling tiny, inexpensive computers and accessories since 2012. But while Raspberry Pi hardware has become popular with enthusiasts and home users, the organization’s single-board computers were originally targeted at students, educators, and makers. The new Raspberry Pi 400 is something different: a full-fledged computer stuffed inside a keyboard. It’s basically a modern take on the Commodore 64 and similar personal computers from decades past. A key difference? The Raspberry Pi 400 is incredibly affordable, with prices starting at just $70. http://https://liliputing.com/wp-content/uploads/2020/11/raspberry-pi-400-768x548.jpg Here’s a run-down of the new computer’s key specs: Processor: Broadcom BCM2711 processor (4 x ARM Cortex-A72 CPU cores @ 1.8GHz Memory: 4GB LPDDR4-3200 Wireless: Dual-band WiFi 5, Bluetooth 5.0 Ports: Gigabit Ethernet, 2 × USB 3.0 and 1 × USB 2.0, 2 × micro HDMI ports (up to 4Kp60) GPIO: Horizontal 40-pin GPIO header Storage: MicroSD card slot for OS and data storage Power: 5V DC via USB connector Keyboard: 78- or 79-key compact keyboard (UK and US English, French, Italian, German, and Spanish layouts available at launch) Dimensions: 286 mm × 122 mm × 23 mm And for the old Commodore 64 users out there, here is the comparison between the new Pi 400 and the 1985 Commodore 64 unit. http://https://www.raspberrypi.org/app/uploads/2020/11/C64_comparison-800x373.png |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/03/20 at 04:30:32 5nm chipsets are jest a rolling out right now from both TSMC and from Samsung 5nm plants. The Exynos 1080 is a 5nm chipset, though the exact details on the manufacturing process are not known for now. It will join the growing ranks of 5 nm chips, starting with the Apple A14 but also including the Snapdragon 875, Kirin 9000 and Samsung’s own flagship chipset (the Exynos 2100). Each new week gives us another 5nm announcement, all with ship dates taking place this year. This wave is early, a full half a year early in fact. Yields are very good too, much better than 7nm yields actually so processor costs at retail should drop fairly quickly as scrap costs will be reduced. Volume-wise TSMC is in the lead, having built run after run of Apple chipsets, some Kirin chipsets for Huawei and a few runs of Snapdragon chipsets for Qualcomm. A couple of runs of AMD 5nm multi-layer chiplets are up next on the roster. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/03/20 at 12:25:43 https://liliputing.com/2020/11/arm-unveils-cortex-a78c-cpu-for-always-on-laptops.html ARM unveils Cortex-A78C CPU for always-on laptops http://https://liliputing.com/wp-content/uploads/2020/11/arm-cortex-a78c_01-768x374.jpg This 5nm design was promised to us at mid-summer by ARM for next year ---- it has arrived now and right now any of the phone boys can license this tech and build their own PC chipset. Current design offers are 6 cores and 8 cores --- 8 cores would be a fairly good PC if matched with a good upper end Mali GPU. THIS IS LIKELY WHAT QUALCOMM IS USING IN THEIR X-1 VERSION OF IT. ARM Cortex-A78C has far more cache than other members of the family, and supports up to 8 “big” CPU cores on a single chip, which should offer much better multi-threaded performance than a smartphone-class Cortex-A78 chip, which would top out at 4 “big” cores and 4 “LITTLE” cores based on Cortex-A55 architecture. According to ARM that, combined with 8MB of L3 cache memory and other improvements should lead to better performance for gaming or workloads that involve large datasets. It’s likely that the new CPUs will be a little more power-hungry than their cousins that use a big.LITTLE design. But ARM-powered laptops already tend to get pretty long battery life, so I suspect PC makers (and customers) would be willing to sacrifice a little run time for a significant enough boost in performance. ARM doesn’t actually manufacturer and sell its own processors, so it’ll be up to third-party companies to license the new designs and use them in upcoming laptops or other mobile devices. http://https://liliputing.com/wp-content/uploads/2020/11/arm-cortex-a78c_02-768x416.jpg It is now Nov 11 and Apple is shipping their M1 chipset at 5nm. Overall structure looks a lot like this one. Mediatek has announced new chipsets at TSMC's new low cost lithography stages which are at 6nm, and at 8nm and at 12nm. People say they have the same design rules as 7nm, 10nm and 14nm so all your old designs have a potential new home at the new smaller lithography nodes. Life going out into the future looks very hard if your name is Intel ....... ::) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/04/20 at 15:49:06 https://arstechnica.com/gadgets/2020/11/intel-enters-the-laptop-discrete-gpu-market-with-xe-max/ ArsTechnica is still deep diving Tiger Lake and Intel's Xe graphics, basically telling us that Intel's new hardware isn't much better than the old Intel hardware was, but the AI background cheating stuff is where all the new "Intel Magic" is coming from. In Intel's demonstration, an Xe Max-equipped laptop used Gigapixel AI to enhance a very large, grainy photo seven times faster than a similar laptop equipped with an Nvidia MX 350 GPU could. While that was impressive, we pressed Intel for comparisons to other hardware, which an Xe Max-equipped laptop might more reasonably compete with "in the wild." After a day or so, an Intel engineer got back to us and said we could expect an Xe Max-equipped laptop to complete Gigapixel AI workloads seven times as fast as an MX 350, five times as fast as an RTX 1650, and 1.2 times as fast as a Tiger Lake laptop with Iris Xe graphics alone. Folks, don't allow Intel to confuse their AI tricks with their hardware's real performance, not in your minds in any case. Intel is simply AI cheating on a scale not seen to date even from the loosest of the oriental cell phone builders. So a VERY LOUD Intel buyer beware alarm warning horn blast is still appropriate for Tiger Lake models right now ........ AVOID THE TIGER LAKE GENERATION COMPLETELY --- Tiger Lake is based on a set of AI software tricks that has to be written into your applications which then has to use Intel driver libraries that are out on the web to get to the very latest AI driver tricks to actually use them. Until this happens, it is all smoke and mirrors, smoke and mirrors. =================================================== So, let's take a deep breath here, anything that can give a 7x or 5x increase only in a certain set of tasks might seem like it could be worthwhile if it is done legitimately. But right now it isn't being done correctly and it isn't real AT ALL right now for most commonplace software uses. So AI can be potentially worth something, potentially worth something more than a lithography shrink for that particular set of tasks and for that particular vendor, especially for a company like Intel that simply cannot pull off a lithography shrink in any reality time anywhere near now. BUT, the current Intel secretive black bag "Intel Marketing" trickery is still coming across as despicable and somewhat fraudulent and it seems like outright lying to the purchasing public at times the way they have done it recently ..... Especially when you read this stuff PRINTED IN BOLD TYPE ON THE CARDBOARD BOX after it gets so poorly done going through a machine builder's advertising and sales departments, done so very poorly it can sometimes reads like outright false advertising in motion. http://https://cdn.arstechnica.net/wp-content/uploads/2020/11/encoding-vs-rtx-2080-super.png Note how small the actual hardware improvements are compared to the AI boost given by the software designed to use Intel's AI functions (via Intel's Deep Link Media Encoders). Let's not be totally negative ---- let's end this review on a hopeful note. Ars Technica participated in a briefing at the end of the day and Ars had pressed away at the Intel representatives for more details ...... and Ars found out that honestly most gamers and casual content creators won’t see much benefit to getting a Tiger Lake + Iris Xe MAX laptop over getting one with plain old Iris Xe graphics (standard Intel on board graphics). But that doesn’t mean there aren’t any advantages out there somewhere in this mess. And it’s possible things could get better sometimes out in the future. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/06/20 at 12:21:23 https://www.anandtech.com/show/16214/amd-zen-3-ryzen-deep-dive-review-5950x-5900x-5800x-and-5700x-tested/31 This is a massive Anand Tech review of the first Ryzen 3 "x" processors from AMD that were promised to arrive around Nov. 5th after the Ryzen 3 processors and the "x" grade was announced in early October. It is now Nov. 6th and the Anand Tech review is here as promised. The entire review is 30 web pages long and it is an exhaustive, exhaustive review to say the least (and if you are an Intel fan, the entire thing is totally, totally painful and demoralizing for you to read it all). There is a scroll bar up at the top of each page on this review, as each set of the 30 sets of web pages covers just one of thirty topics or tests ...... So, the angry green She Hulk did catch up with Intel early on in the review and she simply wadded him up so tightly so he could fit into the standard Intel octagonal box along with a suitable brown stinky Intel Marketing crown that was freshly laid down on his bleeding scalpless skull personally by the angry green She Hulk. As last glimpsed, "Intel inside the box" was still struggling a tiny bit, still twitching and weakly squeaking a tiny bit underneath his current stinky brown crown as She Hulk tucked the last flap of the octagonal box flap into place and then she carefully installed the little "Intel Inside" sticker seal over the now security sealed box flap of the (still dripping from the bottom) Intel octagonal container ....... Conclusion: AMD Has Indeed Ryzen To The Top Coming out the other end of this review, it’s hard to believe the extent to which some of AMD’s performance numbers have grown in the last five years. Even within the Ryzen family, we can pinpoint the leaps and bounds by which AMD is now the market leader in pure x86 performance. Let’s start with some headline numbers. +19% IPC Gain is Confirmed at 24% AMD quoted to us a raw IPC gain from Zen2 to Zen3 of +19%. AMD measured this with 25 workloads and both processors at 4.0 GHz, running DDR4-3600 memory. By comparison, we test with industry standard benchmarks at rated clock speeds and JEDEC supported memory, and we were able to accurately achieve that +19% number. http://https://images.anandtech.com/doci/16214/IPC%20Gen-on-Gen3_575px.png Compounding the generation-on-generation gains from a pre-Ryzen era, we’re seeing +114% IPC improvements, and if we look from the original Zen to Zen3, it is a ~41% gain. In real world benchmarks, we saw an average +24% performance gain, showcasing both the increase in IPC and frequency uplift that the Ryzen 5000 parts have. 5.0 GHz Achieved on Ryzen Turbo frequencies are often setup dependent, and because AMD uses a combination of listed turbo frequency and opportunistic boosting, the exact turbo frequency can be hard to nail down. For the top-tier Ryzen 9 5950X, AMD lists the turbo frequency as 4900 MHz for single core loading, however in very standard conditions, we were able to pass that to 5050 MHz. Diving deeper into the AGESA, this processor actually has a ‘maximum frequency’ setting of 5025 MHz. All of our Ryzen 5000 series processors offered +50-150 MHz above the listed turbo showcasing that these parts still have some headroom. http://https://images.anandtech.com/doci/16214/5050_575px.png Overall Impressions of Zen 3 and Ryzen 5000 One of the exciting things about probing a new core is finding out all the little quirks and improvements that they don’t tell you about. It’s been interesting finding out how this core was put together, both from our findings and discussions AMD’s engineers. Moving to an 8-core CCX for this generation was a no-brainer, with an easy up-tick in performance. However it is the changes in the execution units and load/store were a lot of the magic happens – increasing to peak 3 loads/cycle and 2 stores/cycle, splitting out some of the ALU/AGU work, finer grained transitions from decoder/op-cache to the micro-op queue, and pre-fetchers with more bandwidth all help to that. A lot of the instruction improvements, such as lower latency FMA and faster DIV/IDIV is going to scale well as we move into the enterprise and EPYC processors. With AMD taking the performance crown in almost areas it is competing in, attention now rolls over to price. Having $300 as an entry level for this tier is going to sting a lot of users who would rather spend $200 or less – despite AMD having nine out of ten of Amazon’s best sellers, only two of those parts are $300 and up. There’s going to be an early adopters tax as well – one could argue that moving into Q1, when AMD is enabling 400-series motherboards, might be a better inception point for a lot of users. Having said that, with Intel set to launch Rocket Lake at the end of Q1 next year with 8 cores, this sub-$300 market is going to be ripe for any AMD Zen3 APU to come in and take that price bracket. Remember, AMD never launched Zen2 APUs into the consumer market, which might indicate a fast follow-on with Zen3. Watch this space – a monolithic Zen3 APU is going to be exciting. AMD Ryzen 5000 Series Processors Ryzen 9 5950X 16c/32t 3400 4900 64 MB 105 W $799 Ryzen 9 5900X 12c/24t 3700 4800 64 MB 105 W $549 Ryzen 7 5800X 8c/16t 3800 4700 32 MB 105 W $449 Ryzen 5 5600X 6c/12t 3700 4600 32 MB 65 W $299* All things considered, we’re really impressed with what AMD has achieved here. After the disillusionment of years of weaker generation-on-generation performance uplifts from the competition, AMD set a goal to beat the average ~7% IPC year-on-year gain. With +19% IPC on Zen3, Intel has no product to match AMD right now - not even Tiger Lake at 4.8 GHz - and Intel has lost that single-threaded crown. AS SUCH, AMD now wins the total brand crown for GAMING and for content creation as AMD sweeps the top 4 spots before an Intel processor even gets a listing in the rankings. http://https://images.anandtech.com/graphs/graph16220/119145.png http://https://images.anandtech.com/doci/16214/A_EditorChoice_Gold2_575px.png Note please that AMD can easily break 5.0 ghz at only 105 watts of power, while Intel takes 140 watts or more to do the same upper level ghz speeds. Also note that this is all done at a bog standard TSMC 7nm lithography level. Also note that with the brand new just breaking AMD 5nm chiplets that are currently being run off at TSMC there will be BRAND NEW WAVES OF AMD PROGRESS TO FOLLOW QUICKLY AFTER THIS ONE, each new wave lapping up significantly higher on the beach and lapping out fairly quickly too. =================================================== Look for new ever higher AMD progress waves to hit the beach scheduled roughly every six months apart until every Intel marketing sandcastle is completely leveled and totally washed away (this will take place easily inside the next calendar year no less) ....... TO PROTECT THEIR CASTLE WALLS from scaling ladders, Intel has added more height to the speed crenellations, raising them from 5.0 to up to 5.6 ghz tall ....... at 160 watts of power consumed and even bigger heat pipe cooling fins and fans to match. This makes no difference to the AMD groundswell waves that are coming in, waves that plow into the Intel built castle walls down low, eroding sand from underneath their foundations, not by going over the tops of Intel's best hyper stacking over voltage efforts. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/09/20 at 17:22:35 http://linuxgizmos.com/risc-v-based-allwinner-chip-to-debut-on-13-linux-hacker-board/ http://linuxgizmos.com/files/allwinner_riscvsom_cnx-sm.jpg Allwinner is building a <$13 hobby board using a RISC-V chipset. Alibaba’s T-head subsidiary amd Allwinner have produced a single-core, RISC-V-based XuanTie C906 processor with MMU that will appear on a sandwich-style, Linux-driven, $12.50 Sipeed SBC due in two months. In July, 2019, Chinese tech giant Alibaba Group surprised the industry with the most powerful RISC-V architecture SoC design to date: a 16-core, 2.5GHz XuanTie 910 (XT 910). Now Alibaba’s RISC-V-focused T-Head subsidiary is collaborating with Allwinner to introduce a single-core, 1GHz XuanTie C906 (RV64GCV) RISC-V processor designed to run Debian Linux. A Sipeed dev kit will arrive in the coming months for $12.50 that incorporates an Allwinner compute module built around the XuanTie C906. Note: better single board computers exist, notably the Raspberry Pi 4.0 units set the minimum acceptable end of all single board computing for all rational beings, which makes this one still a complete non-player. This exists to show that a lot of Orientals are unhappy with USA and Britain still controlling their uses of computing. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/12/20 at 05:48:05 https://www.tomshardware.com/news/intel-now-content-to-match-tsmcs-process-technology Intel Now Content To 'Match' TSMC's Process Technology By Lucian Armasu December 10, 2019 This Bob Swan statement is a year old now, but it mostly rings in as completely true as what Intel is doing. If it wasn't for black bag cheating, Intel HAS NO PROGRESS TO OFFER as the Intel 10nm only works with a bunch of black bag tricks and Intel 7nm doesn't work at all right now. At a fireside chat with Credit Suisse at their 23rd annual technical conference, Intel’s CEO Bob Swan said that its 7nm process is expected to match TSMC’s 5nm process. He also noted that Intel's 5nm process is also expected to match TSMC’s 3nm process. Neither of these things worked out, however. What Swan didn’t mention is that Intel is no longer in the lead in terms of process technology and that its 7nm process is expected to arrive about a year or so later, in 2022, compared to TSMC’s 5nm which will produce device chips by the second half of 2020. Intel's Recent Process Roadmap When Intel announced the 22nm Tri-gate (FinFET) process, it was more than a generation ahead compared to TSMC and other third-party foundry competitors. For one, it was on a smaller 22nm process compared to others who were just moving to 28nm/32nm process nodes. And second, the switch to FinFET alone granted its own generational boost in performance and efficiency. Intel’s process leadership was uncontested for years after that. The only exception was in mobile chips, where its 22nm FinFET Atom chip could barely match the latest high-end Arm chips on a 28nm planar node, and at a higher chip cost to boot. It’s why Intel eventually attempted to license the Atom design to Chinese fabless semiconductor companies so that they build their own “Atom” chips more cheaply on TSMC’s 28nm process. However, that didn’t work out either, as few device makers were interested in making the trade-off compared to the Arm chips they’ve come to rely on for each device launch. Intel then switched to 14nm, and that wasn’t a very smooth ride either. The company experienced some delays with Broadwell chips, which were the first to use the 14nm process. Intel also ended-up replacing the consumer Broadwell generation on the market with Skylake rather quickly. The reason the switch to 14nm was quite bumpy, too, was because Intel set out to increase the chip density by 2.4x, which ended-up a little too difficult to implement, but the company succeeded in achieving that eventually. However, instead of taking this lesson to heart, Intel attempted to increase the density even more aggressively with the 10nm process, by 2.7x. After years and years of delays, the company recently admitted that the goal was too aggressive for the company. This is why for its move to 7nm EUV, Intel will scale back the density increase to 2.0x. The switch to an extreme ultraviolet (EUV) lithography process is difficult enough as it is. It’s also Intel’s first attempt to implement EUV, following in Samsung and TSMC’s footsteps. From a Leader to a Follower? Although the claims of Intel’s CEO are meant to reassure both customers and partners that the company is “still” as competitive as TSMC, this statement doesn’t come from a position of leadership in this aspect. Even if what Intel claims is true and the process is indeed more or less equal performance, efficiency, and density-wise, the fact is that devices with chips made on TSMC’s 5nm process are expected to come out in the second half of 2020. Those devices will include the next-generation iPhones, too. Meanwhile, Intel has recently said that it will start launching 7nm chips sometime in 2021. Intel has also fallen behind AMD in terms of having secure processors and firmware, going by the number of recent vulnerabilities (surpassing AMD 15:1). The company also says that it’s no longer interested in large market share, coincidentally in a time when AMD’s Zen 2 chips are expected to make the biggest dent in Intel’s sales, both in the consumer and data center markets in more than a decade. Although none of this means Intel still isn’t in a much stronger position both brand and financially-wise compared to AMD and TSMC and that it can’t recover, it’s still a trend that doesn’t do Intel any favors and that Intel will have to reverse this trend before it’s too late. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/12/20 at 09:12:54 https://liliputing.com/2020/11/samsungs-first-5nm-chip-is-the-exynos-1080-for-mid-range-phones.html Samsung’s first 5nm chip is the Exynos 1080 for mid-range phones http://https://liliputing.com/wp-content/uploads/2020/11/exynos-1080_04.jpg The Exynos 1080 features: 1 x ARM Cortex-A78 CPU core @ 2.8 GHz 3 x ARM Cortex-A78 CPU cores @ 2.6 GHz 4 x ARM Cortex-A55 CPU cores @ 2 GHz Mali-G78 MP10 graphics Integrated Neural Processing Unit Integrated 5G mode The new Samsung Exynos 1080 processor is an octa-core processor deigned for upper mid-range smartphones, but Samsung says it will bring a 50-percent boost in single-core performance and up to a 2X improvement in multi-core performance, which could make this mid-range chip competitive with some processors designed for high-end phones and tablets. Samsung’s new chip is also the company’s first manufactured using a 5nm EUV process. We are looking at some significant throughput increases for Samsung 5nm chipsets ---- up to 2x multi-core and 50% for single core. Plus wafer area savings that says each chip can have more memory and more layers and more other goodies inside, basically for no additional wafer silicon cost. And at better yields than 7nm as well, lowering the scrap costs. More chips per wafer and far lower scrap rates means SUBSTANTIAL COST SAVINGS to the builders, something nobody has seen for a while over there in silicon land. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/12/20 at 11:01:20 RUMOR TIME AGAIN ....... Some facts, some rumors -- all mixed together in the wok just to improve the juicy flavorful sauce ....... TSMC is building a 3.5 Billion $$$ Arizona USA facility to do high volume 5-6nm and 3-4nm lithography at up to 14 layers deep. Apple is joining TSMC Global with a big arsed stock swap. This is why they let the ARM deal go on past them to NVIDIA. Intel is going to be the largest customer of the Arizona facility, even larger volume-wise than Apple and AMD added together. Even MediaTek has neural processors hidden inside their most current designs. 3x-5x upticks in overall processing speed will come some from the lithography shrinks going down to 5nm and the rest of general progress will mostly come from AI integrations. Microsoft needs to use Linux to utilize these new ARM features and speed levels, existing MS product lines simply can't run that stuff at full speed as MS is too tied down to the x86 past to take in all the changes. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/17/20 at 11:30:44 https://liliputing.com/2020/11/apple-mac-with-m1-chip-review-roundup-blazing-fast-energy-efficient-with-some-frustrating-limitations.html 5m Apple M1 chip devices are shipping now for the holidays, and they are SIMPLY BETTER in the light laptop range than x86 based laptop chipsets from either Intel or from AMD. Not in all cases, nor in all particulars, but M1 is better in the sub-40 watt range than x86, generally speaking. For the first shot out of Apple's gun, this is a compelling first entry as Apple's first home grown chipset cleanly hits the mark it was aimed at. To say it totally guts Intel Tiger Lake is an understatement ...... Apple will then build them some bigger and stronger chipsets for desktop uses, count on it. Out of the gate, AMD and Intel are being humbled by this processor. This will prompt a fast response from AMD, but Intel has no new bullets ready to load in any weapon that they have built out already ..... (sad but true, Intel will be mostly silent for the next year in this conflict) Intel cannot count on their "Intel BS Black Hat Marketing Games" with AI to work against Apple as Apple has totally OS integrated all such stuff already and is currently sitting on an 11 TRILLION Operations per Second built in AI Neural Engine that has Apple Standards based support already built into all of Apple's softwares. AI is expected by the phone boys, it is not a "secret weapon" of any kind .... The ARM based phone boys will shoot back sooner and more effectively than Intel can in this new laptop conflict range. http://https://liliputing.com/wp-content/uploads/2020/11/m1-info-1-768x432.jpg http://https://liliputing.com/wp-content/uploads/2020/11/m1-cpu-1-768x432.jpg Here are some of the many many first reviews of Apple’s new Macs with M1 chips: Macbook Air with M1 Business Insider CNBC CNN Engadget Fast Company iMore Gizmodo Laptop MacWorld Mashable TechCrunch Tom’s Hardware The Verge Wired WSJ MacBook Pro with M1 Laptop MacWorld MKBHD (YouTube) Pocket Lint TechCrunch The Verge WSJ Mac Mini with M1 AnandTech TechCrunch The Verge General M1 reviews Ars Technica Six Colors AnandTech’s Mac Mini review also has a deep dive into some of the chip details that Apple hasn’t publicly disclosed. For example, the Apple M1 processor’s four high-performance “Firestorm” CPU cores seem to be clocked at 3.2 GHz, which is a little higher than the 3 GHz speed of the Apple A14 chip used in the latest iPhones and iPads. Power consumption seems to range from 4.2 watts at idle to about 27 watts under heavy workload. So power consumption is roughly similar to what you’d expect from a computer with a 20-24 watt TDP, as measured by other chip makers. =================================================== But there are a few other good reasons Apple building its own ARM based chips could be good for the PC market more generally. First, it could spur other chip makers including Intel, AMD, Qualcomm, and MediaTek to up their game. Second, it could give companies like Google and Microsoft a reason to try developing their own chips in-house to very tightly integrate hardware and software as Apple has just done. And third? It gives developers of popular applications a reason to port their software to run natively on devices with ARM-based processors. Any of these happening is a blow to Intel, make no mistake about it. The age of "Intel on x86 total dominance" is jest about over. Intel will become relatively a smaller and smaller player with many many new competitors constantly popping up and egging each other on continuously, starting right now. today. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/18/20 at 18:19:48 https://liliputing.com/2020/11/lilbits-porting-apps-to-arm.html Rapid progress adapting software to use the Apple M1 Chipset and all of its capabilities. Apple has published AI specifications to use the 11 trillion operations per second Neural Processor that is part of all M1 chipsets. This isn't a secret and it isn't the Black Ops thing like Intel is trying to do it. As such, the future belongs to Apple's relatively open system of calls and requirements. Being relatively open, Apple puts pressure on Intel to come in out of their Black Ops secrecy and to publish what their AI requires out of an OS or out of an individual software program to actually use the AI boosts that are in the Intel processors and the large number of associated daughterboard chipsets. Until this happens, Intel customers are paying a significant price bump to buy a complex daughter board mounted AI system that right now only Intel can utilize for certain benchmark tests ........ The list of companies jumping on the Apple Silicon bandwagon is growing. Google just released a version of Chrome for Apple’s new computers with M1 processors, and Mozilla is developing a version of Firefox that will support the new chips as well. Meanwhile, Adobe’s recent move to release a beta version of Photoshop for both macOS Big Sur devices with Apple M1 chips and Windows on ARM has had an unintended side effect – at least one developer who has loaded Windows 10 onto an old Microsoft Lumia smartphone can now run Photoshop on a phone. Google releases Apple Silicon-optimized version of Chrome for Mac Google releases a new version of Chrome for macOS that’s optimized for the Apple Silicon chips in the new MacBook Air, MacBook pro, and Mac Mini. Pre-release Firefox for Apple Silicon [Mozilla] And Mozilla is working on a version of Firefox that will do the same. Tab throttling and more performance improvements in Chrome 87 [Chromium Blog] Chrome 87 will actively manage your browser tab resource usage to balance performance and power usage – extending battery life while maintaining responsiveness. Back/forward cache will also improve page load speeds when you hit the back or forward button. Google Pay reimagined: pay, save, manage expenses and more [Google] Google unveils new Google Pay for Android and iOS (US-only at first), with money management/insights features, support for person-to-person, group, and business payments and collections, and more. Plex mobile-first bank accounts are coming next year. Meet the Microsoft Pluton processor – The security chip designed for the future of Windows PCs [Microsoft] Microsoft introduces Pluton security processor. It will eventually replace TPM in upcoming PCs with AMD, Intel, and Qualcomm chips, offers protection from speculative execution attacks, and can receive updates through Windows Update. Huawei to sell Honor brand to consortium of agents and dealers in bid to save its supply chain [Reuters] Huawei’s been under pressure due to US trade sanctions that have limited the company’s ability to acquire components for its devices. It’s unclear whether this new ownership structure will help the Honor sub-brand at all Purism’s Librem 5 Linux smartphone is now shipping [Linux Smartphones] Three years after announcing plans to build a Linux smartphone with an emphasis on free and open source software, privacy, and security, Purism is now starting to ship the mass production version of the Librem 5 to customers. http://https://liliputing.com/wp-content/uploads/2020/11/pshop-768x576.jpg Photoshop on a Lumia phone running Windows on ARM [@imbushuo] Not only does Photoshop now run on Mac and Windows computers with ARM-based processors… but folks who have shoehorned Windows on ARM onto phones for some reason or other can try using it on those too. Also, screen rotation is now working on that Lumia phone with Windows on ARM [@imbushuo] Speaking of running Windows 10 on phones, now it looks like you can do that and easily switch between portrait and landscape. https://www.codeweavers.com/blog/jwhite/2020/11/18/okay-im-on-the-bandwagon-apple-silicon-is-officially-cool This is the Code Weavers guy burbling in Apple-speak that he's got it working already. Okay, I'm on the bandwagon - Apple Silicon is officially cool by Jeremy White The new Apple Silicon-based Macs have dropped and the reviews are very positive. They've delivered faster machines that use less power, and they run CrossOver 20 brilliantly! We have a range of the new systems on order. But we got impatient and discovered that our local Best Buy had the cheapest Macbook Air in stock, so we bought it and loaded CrossOver 20.0.2 onto it. We also installed the beta version of Big Sur 11.1, because we know it has some critical fixes to Rosetta. After we did that, we were able to fire up CrossOver and install and run a wide range of Windows applications. We got Quicken to work: the desktop version of Among Us, where you can actually use your mouse, and we were even able to have a satisfying session of Team Fortress 2: That's incredible when you consider that we're on literally the cheapest Apple Silicon device you can buy - one that gets thermally throttled and is missing a GPU core. I can't tell you how cool that is; there is so much emulation going on under the covers. Imagine - a 32-bit Windows Intel binary, running in a 32-to-64 bridge in Wine / CrossOver on top of macOS, on an ARM CPU that is emulating x86 - and it works! This is just so cool. Now it isn't perfect; Team Fortress 2 showed some lag. I think we've got some work to do on that front. But I can't get Brian to stop playing Witcher 3... Cheers, Jeremy |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/21/20 at 21:35:40 https://youtu.be/xNNTXd85VXo This quick video review is worth watching. OK, same-same model comparison showing that the state of art Intel processor Mac can't keep up on any head to head test against the much lower current draw M1 Apple Mac. The Intel based Mac gets HOT, blows a whole lot of hot fan air ..... and the cool running M1 Apple Mac does not even kick its fan on, even though it is noticeably faster getting the tests done (and yes, it is emulating to do that on some of the tests, but it is emulating quicker than Intel can even run naturally). "The fans have not kicked on once with the M1 Mac during this entire test series" =================================================== I would say AMD needs to change its focus away from Intel and focus on their real competition, which is the AI equipped (11 trillion ops per second neural processing) ARM based computer chipsets. Yes, with the Apple M1 Mac simply being the first of these ARM 5nm chipsets to ship to the buying public that is complete with a decent neural processor. ARM has all this stuff sitting there ready to license and trust me, the phone boys will buy into it. Focusing anything on Intel is wasting precious development dollars and critical development time as Intel is doing ABSOLUTELY NOTHING meaningful right now, completely zero progress and pursuing totally zero real world wins. Lastly, on the subject of AI and all the neural processor stuff, Apple does it very very well, much much better than Intel can. Next, the larger "outside of Apple" software world already has all the neural libraries and AI drivers and files and support stuff already out there to make their Apple related products sing right along, whereas Intel is having a very hard time even getting started outside of Intel itself. (it is very tough to keep all the lies and BS secret games going if others are involved ......) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/24/20 at 00:51:01 https://www.ftc.gov/sites/default/files/documents/cases/091216intelcmpt.pdf In 2009 Intel was enjoined by the FCC specifically to not do the things that Intel is currently doing again, big time. Trump's administration may have turned a blind eye to this activity by defunding the FCC, but Biden's administration is likely not to ignore Intel's activities any further. If you open the pdf and read it, it is EXACTLY what Intel is up to all over again ..... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/24/20 at 13:41:45 Intel busted a gut recently, lying profusely to set themselves up as the ultimate benchmark for "gaming performance" Now, everyone else is now using Intel to prove they are better than "x86 as was in the past" and by golly they ACTUALLY MEAN it this time around ....... and they can prove it by like to like same model comparisons. Intel blows a lot of hot air around, and performs poorly compared to the new stuff from AMD and APPLE. =================================================== Intel is now trying to brag that they made up the specs for a 10 core Intel processor (but they were unable to get their first runs to work right as Intel had to laser off several non-functional cores in testing) thus making up some dodgy poorly performing 8 core chipsets with some odd chipset designations. The successful ones that came out as 10 core chipsets didn't perform one whit better than the dodgy 8 core cut downs, and actually didn't do much better than their 6 core predecessors. Faced with ever increasing post sorting scrap rates, Intel would have to make up a 14 core layout to yield 10-12 "meets spec" cores, or else they will do the standard old style Intel reaction plan yet again, which is to jack the final spec limits wide open so that everything "passes" Intel final test bench inspection as it is is. LOOK FOR THE OLD "YOU CAN'T DO ANYTHING TO TEST BENCH THE PERFORMANCE SPECS ON YOUR OWN INDIVIDUAL CHIPSET" MISLEADING INTEL BULLSHITE TO GET TROTTED OUT AGAIN TO COVER INTEL'S DECISION TO SHIP OUT THEIR IN-PROCESS SCRAP ON TO THEIR END USERS. =================================================== Look to see AMD to make themselves up a separate neural chiplet, large, powerful, distinct and easily produced by the wafer load. Look to see the AMD APU "all on the same chip silicon" go away at 5nm and smaller, as using AMD's chiplet construction methods means FAR LESS SCRAP and waste are created at each new generation time. Look to see other chipset manufacturers start to copy the chiplet idea, as it leads to faster six month generation cycles with significant improvement levels at each level. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/26/20 at 14:22:55 https://www.cnx-software.com/2020/11/26/rockchip-rk3588-specifications-revealed-8k-video-6-tops-npu-pcie-3-0-up-to-32gb-ram/ Rockchip RK3588 specifications revealed – 8K video, 6 TOPS NPU, PCIe 3.0, up to 32GB RAM Rockchip RK3588 is one of the most anticipated processors for the year on this side of the Internet with the octa-core processor features four Cortex-A76 cores, four Cortex-A55 cores, an NPU, and 8K video decoding support. The roadmap shows an expected launch date in Q3/Q4 2020, but sadly the release date will be pushed back in the future. Having said that, the Rockchip Developer Conference (RKDC) is now taking place, and the company has put up a poster that reveals a bit more about the processor. Rockchip RK3588 specifications That means we now have some more detailed Rockchip RK3588 specifications available now. CPU – 4x Cortex-A76 and 4x Cortex-A55 cores in dynamIQ configuration GPU – Arm Mali “Odin” MP4 GPU AI Accelerator – 6 TOPS NPU 3.0 (Neural Processing Unit) VPU – 8Kp60 video decoding support, 8Kp30 encoding support Memory I/F – LPDDR4x/LPDDR5 up to 32GB Storage – eMMC 5.1, SDIO, SATA 3.0 (multiplexed with PCIe 2.0) Video Output Dual HDMI 2.1 / eDP up to 8Kp60 or 4Kp120 Dual DisplayPort up to 4Kp60 Dual MIPI DSI output Up to four independent displays Camera – 48M (2x 24M) ISP with HDR and 3D NR support; multi-camera input Audio – Microphone array support Networking – Dual Gigabit Ethernet USB – 2x USB 3.1 Type-C, 2x USB 2.0 PCIe – 4-lane PCIe 3.0, and 3x PCIe 2.0 (multiplexed with SATA) Manufacturing process – 8nm LP RK3588 Rockchip-Developer Conference 2020 The company will provide support for Android, Linux, and a “domestic OS”. We should note that the GPU has changed from “Natt” family to “Odin” family. Rockchip is unable to disclose the GPU name as it’s a new GPU family that has yet to be announced by ARM. We were informed that Rockchip designed its own NPU IP for RK3588, and did not use a third-party design like VeriSilicon NPU IP. Rockchip will likely eventually adopt whichever NPU IP vendor winds up the winner out of the first 3-4 candidates out here at this time. Not enough is in use to call out a "best vendor" at this point in time. RK3588 specifications are pretty impressive, and the processor will be found in Arm computers, smart displays, edge computing & AIoT solutions, Arm servers, high-performance tablets, network video recorders, virtual reality headsets, and applications requiring multiple cameras and displays. We should however note that RK3588 sadly lacks UFS support, and relies on the slower eMMC 5.1 flash interface for storage. The launch is now scheduled for Q2/Q3 2021. Rockchip represents the typical low end ARM SOC lapping up into something that can easily run a chromebook, a Linux laptop or an "intensive use" tablet with keyboard. This also acts to raise the performance bar for Mediatek's low end processors, a general low end competitive move upwards which is always a good thing. See this Mediatek roll up to show up in AMAZON Fire Tablets next year. Notice that ARM provides the designs of neural processing units (AI cores) which all of these chipsets now incorporate as Android provides for this use model (and has for 5 years now). All of the phone boys have neural processors and use the newer smaller lithography cores available to them now as well as pre-proven out canned computer processor layouts direct from ARM that they can have build in volume at TSMC. What Intel struggles to incorporate now is NOT NEW --- the PC people are all playing catch up to the phone boys on ARM technology uses yet again. AI gives full multiples of better software performance but only where it is fully implemented in the surrounding larger software world. AI will be important in the next few years as hardware increases are now down to the <20% level per generation, a situation which says the big bang for the buck comes from the 6 to 11 trillion instructions per second neural processors, not the standard lithography shrinks. AMD is behind in this AI race, and is playing catch up going forward. AMD has arguably the best AI partner out of the lot for their AI partner. =================================================== https://liliputing.com/2020/11/lilbits-windows-on-apple-silicon-android-apps-on-windows.html Apple Silicon is fast. Macs with Apple M1 chips outperform just about every other Tiger Lake laptop on the market by >20% in most benchmarks, while still consuming less power. Even Mac apps that were designed to run on x86 chips run faster on the new Macs than they do on equivalent models with Intel processors. But what if you want to run a different operating system? Apple’s Craig Federighi recently told Ars Technica that it’s up to Microsoft to decide whether to support Apple’s new hardware. Previous-gen Macs with Intel processors had a feature called Boot Camp which allowed users to install Windows on their Apple computers and dual boot, choosing which operating system to run at launch. Right now Apple M1 has a slamming lead over any laptop chipset out there. Ranking runs Apple M1 up by 25% to either AMD or Intel. Intel has NOTHING in the pipeline for the next 1 1/2 years while AMD can put out a competitive laptop product as soon as TSMC runs their lots of chipsets. There will be 1-2 Apple vs AMD bouts before the ARM phone boys begin selling their 8 core A-78c product, which may bump past AMD for a short period of time. Then the entire match up becomes "who can implement their AI better and cover more of their general software space" while doing so. Apple has the advantage here as they have already put their support out there and ENFORCED all of Apple software space to switch over to it. http://https://appleterm.com/wp-content/uploads/2020/11/dell-xps-vs.-m1-macbook-air-side-by-side-rect.jpg |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/01/20 at 15:09:46 https://www.tomshardware.com/news/huaweis-24-core-7nm-kunpeng-920-cpu-allegedly-outmatches-core-i9-9900k-in-multi-core-performance Another vendor lines up to kick Intel i9-9900K's ass Wow, a 24 core 7nm ARM Neoverse based Huawei Kunpeng 920 mainframe chipset swinging a AMD Radeon 520 mobile graphics card (Radeon 520 is a dedicated entry-level graphics card for laptops that was released in 2017). Putting these pieces of 5 year old tech together makes up a new "old style" 7nm processor rig that can kick Intel i9-9900K's 14nm ass right now, today. (to be sure, the Intel 14nm base tech is over 10 years old at this time, so this shouldn't be surprising) http://https://cdn.mos.cms.futurecdn.net/YhaS2ZHLVKYQLpLzT4Sc9g-970-80.jpg Getting perspective, it is a Chinese Government sponsored 24 core 7nm ARM Neoverse mainframe chipset that was a current design back in 2019 that has been combined with a 1917 era AMD Radeon 520 graphics chipset tech that China got from ARM back when it was a kosher US government sanctioned (pre-approved no less) open trade deal. This isn't ARM's best most current tech nor is it run on the current 7nm +++ lithography. It runs on older (used) 7nm lithography equipment that China recently bought used (legally) at the end of its service life. But it still swings a whole lot of cores (24) and equipped with some not terribly moldy AMD graphics so that in total it can totally kick Intel's i9-9900K antique 14nm technology butt once again. Add Huawei's Kunpeng 920 Pangu to the list of computers competing head to head with Intel ........ The Kunpeng 920 3211K in particular has 24 ARM Neoverse mainframe cores that max out at a mainframe under stressed 2.6 GHz. Huawei pairs the processor with 8GB of SO-DIMM memory, a 512GB Samsung SSD and AMD’s Radeon 520 mobile graphics card. Huawei tailors the Kunpeng 920 Pangu to government and enterprise markets, meaning the system is equipped with China's homemade Linux Unified Operating System (UOS). User expansion and customization on the Pangu is close to zero. The Kunpeng 920 3211K is soldered to the motherboard and doesn't support other graphics cards. The UOS is cemented into the PC so you can't install Windows on it either. We suspect you may be able to upgrade the memory or SSD, but that's about it. The purported images of the Pangu show three USB Type-A ports, one USB Type-C port and a single 3.5mm headphone jack in the front of the case. There is also room for an optical drive. The rear of the case holds four USB Type-A ports, one Ethernet port, three 3.5mm audio jacks and a D-Sub port. IThome's report states that the Pangu comes with a 23.8-inch monitor with a resolution of 1920 x 1080 and 70% NTSC color gamut. Some perspective, the same sort of Neoverse cores can be bought from Ampere at counts of 32, 80, and 128 cores so the tech is out there if it becomes more popular over time. ARM itself is selling a 5nm version of the same stuff as the old 7nm node has greater scrap rates than the new 5nm node, with 5nm getting better and more efficient all the time. 3nm is doing well in development testing, so look to see it progress right along just as soon as Apple needs a performance boost to remain "distinctive". |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/02/20 at 07:43:46 https://liliputing.com/2020/12/qualcomm-snapdragon-888-promises-faster-cpu-and-gpu-reduced-power-consumption.html Qualcomm Snapdragon 888 promises 25 percent boost in performance, efficiency (among other things) http://https://liliputing.com/wp-content/uploads/2020/12/sd888_01-768x432.jpg First up, the new processor is Qualcomm’s first 5nm system-on-a-chip, which the company says brings improvements to performance and efficiency. Second, it’s the first processor to incorporate ARM’s Cortex-X1 technology, with the Kryo 680 octa-core CPU featuring: 1 x 2.84 GHz highest-performance Cortex-X1 “supercore” 3 x 2.84 GHz high-performance Cortex-A78 cores 4 x 1.8 GHz efficiency Cortex-A55 cores All told, Qualcomm says you can expect a 25-percent improvement in overall performance, and a 25-percent reduction in power consumption. But the CPU is just one component of the Snapdragon 888 processor. Qualcomm says the new Adreno 660 GPU also brings a 35-percent boost in graphics and a 20-percent reduction in power consumption, the Spectra 580 image signal processor offers 35-percent faster performance and allows you to use a smartphone camera to capture up to 120 pictures per second at resolutions of 12MP, the Hexagon 780 AI processor brings significant improvements as well, and the new integrated Snapdragon X60 5G modem brings support for download speeds as high as 7.5 Gbps and upload speeds up to 3 Gbps using Qualcomm’s 3rd-gen 5G modem technology. The company says its AI Engine delivers up to 26 TOPs of performance. And the company has also brought a Hypervisor to its mobile chipset for the first time, allowing for quick switching between operating systems or allowing for apps to be isolated from the base operating system for privacy or keeping work/home profiles completely separate, for example. Other technology baked into Snapdragon 888 includes a FastConnect system with support for WiFi 6 and Bluetooth 5.2, Quick Charge 5 technology that will allow you to take a phone from a 0 to 50 percent charge in as little as 5 minutes, and support for Qualcomm’s 3D Sonic fingerprint solution. Qualcomm is shipping 5nm chipsets in volume. They are VERY GOOD 5nm chipsets. Qualcomm is rocking 26 Trillion Operations Per Second of pure AI performance, Dude, that jest makes Intel and Apple weep bloody tears, now doesn't it? I told you them durn phone boys were gonna wanna play hard ball in this 5nm ball game ...... They got their sharp metal edged hockey sticks out and got all them narsty metal spikes on their helmets and shoulder pads too. Blood on the grass, Dude, steaming piles of spilled guts too -- yep, you betcha ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/05/20 at 22:51:53 Intel victories and setbacks time again ... https://www.tomshardware.com/news/intel-announces-delay-to-7nm-processors-now-one-year-behind-expectations So, what is going on with Intel? Seems likely another 2 years of nothing much will transpire until sufficient bulk 6nm production space is available from TSMC for TSMC to run all of Intel's new chiplet designs. Intel CEO Bob Swan said the company has identified a "defect mode" in its 7nm process that caused yield degradation issues. As a result, Intel has invested in "contingency plans," which Swan later defined as including using third-party foundries. The company will also use external third-party foundries for its forthcoming 7nm Ponte Vecchio GPUs, the company's first graphics chips. Ponte Vecchio comes as a chiplet-based design and Swan clarified that production for some of the chiplets (Intel calles them tiles) will be outsourced to third parties. Swan noted the GPUs will come in late 2021 or early 2022, portending a 3 year delay beyond the original schedule for a 2021 launch in the exascale Aurora supercomputer. DOE confirms Intel's 3 years of failures & delays https://www.datacenterdynamics.com/en/news/doe-confirms-aurora-delayed-frontier-will-be-first-exascale-supercomputer/ Argonne's Aurora system was originally expected to be the first exascale supercomputer - a system capable of performing more than a billion billion operations per second, ie one exaflops. However, Aurora has been delayed indefinitely due to ongoing Intel manufacturing issues. Delaying Aurora – US Department of Energy Industry observers have expected Aurora to be delayed ever since Intel announced in July that it was struggling to develop 7nm chips, and would delay its Ponte Vecchio GPU by at least two years. With Aurora planned for mid-2021 and featuring GPUs that were pushed to the end of next year or into 2022, a further larger delay on the supercomputer seemed likely. But Intel kept the possibility open that it might manage to hit its deadline by relying on rival chip foundries. We now know that's not going to happen. Department of Energy Under Secretary for Science Paul Dabbar confirmed to insideHPC that Aurora was behind schedule, and that the 1.5 exaflops Frontier supercomputer will launch first. As it uses Epyc CPUs and Radeon Instinct GPUs, both from AMD as there are no concerns about its chip roadmap. Dabbar said that the DOE was "in discussions with Intel" about Aurora, and was looking at "different options," but claimed that he had "no degree of confidence" that the one exaflops system would launch not too long after Frontier. It is unclear how severe Intel's 7nm production problems really are - Intel had previously suffered years of delay on its 10nm development, and in both cases Intel has consistently provided misleading information to the media about how long their delays would last. Looks like 3 years to never, boys and girls, 3 years to never ....... DOE has gone forward with AMD and the Epyc CPUs and Radeon Instinct GPUs for Frontier, the first American Exaflop computer, cancelling the Intel order in disgust due to Intel's repeated lies and total non-performance. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/07/20 at 07:37:01 https://liliputing.com/2020/12/report-apple-is-gunning-for-intel-in-the-desktop-processor-space.html We already see that the older 7nm Neoverse ARM 24 core CPU (Huawei's Kunpeng 920 design) from 4 years ago is still able to kick Intel's ass today. We should not be surprised that current modern designs from APPLE are going to do the same Intel ass kicking with the 5nm designs being most of four years more modern compared to the older 7nm Neoverse ARM 24 core designs from Huawei. Even if all this information is accurate, it’s still a little vague on details and Apple’s plans could change in the coming months. But here are a few highlights from the Bloomberg Apple report that was cited: According to Bloomberg, Apple could release new chips as soon as spring, 2021, with more processors coming from Apple in the fall. Next-gen MacBook Pro laptops and iMac all-in-one desktops could be powered by chips with as many as 16 high-performance CPU cores and 4 efficiency cores (although Apple may also decided to go with 8 or 12 performance cores for the 2021 models instead). Apple is said to be developing chips with as many as 32 high-performance CPU cores for high-end desktop computers. The company is also developing high-performance GPUs with 64 or 128 dedicated graphics cores. This may be an optional upgrade for high-priced, high-power systems. By comparison, the Apple M1 processor that powers the current-gen MacBook Air and 13 inch Macbook Pro is an 8-core processor featuring 4 high-performance cores and 4 efficiency cores, and it still trounces Intel’s latest laptop chips in most benchmarks. ONCE AGAIN, AMD needs to be benchmarking this sort of stuff, not focusing on Intel any further as Intel now has an entire lineup of phone boys forming up to take their turns abusing Intel's aching buttocks with their steel toe ass kicking boots. Up your game, AMD. Kicking up some more on a dead Intel donkey's butt isn't nearly enough for you to be accomplishing for the next calendar year. Dial in some AI, get yourself ready for the next big contest ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/09/20 at 12:21:00 https://www.notebookcheck.net/AMD-essentially-confirms-5-nm-Zen-4-CPUs-and-advanced-node-RDNA-3-GPUs-will-launch-by-2021-in-latest-corporate-presentation.483288.0.html We have noted that AMD's current release schedule was rolled forward by six months, so nothing on the attached page (or the articles beneath that page) should be a total surprise to anyone. What is encouraging is that the 5nm node is indeed currently rolling out like gangbusters and that the 3nm node which follows the 5nm node (the one that is currently in testing) is supposedly doing very well also. We calmly expect AMD to meet all expectations with real hardware, on time as stated. ...... or sooner. The fact that Intel can't ever seem to do so does not lessen our expectations that AMD and ARM will roll out on time and on spec ...... https://www.tomshardware.com/reviews/cpu-hierarchy,4312.html CURRENT GAMING PROCESSOR RANKINGS 1080p Gaming Score 1440p Gaming Score CPU Cores/Threads Base/Boost Watts Ryzen 9 5900X 100% 100% Zen 3 12/24 3.7 / 4.8 GH 105W Ryzen 9 5950X 99.77% 99.38% Zen 3 16/32 3.4 / 4.9 GHz 105W Ryzen 7 5800X 97.22% 99% Zen 3 8/16 3.8 / 4.7 GHz 105W Ryzen 5 5600X 96.90% 95.30% Zen 3 6/12 3.7 / 4.6 GHz 65W Intel Core i9-10995 88.97% 95.30% Comet 14nm 10/20 3.7 / 5.3 GHz 125W Intel Core i9-10855 87.36% 94.52% Comet 14nm 10/20 3.6 / 5.2 GHz 95W Core i7-10700K 84.39% 92.05% Comet 14nm 8/16 3.8 / 5.1 GHz 125W Intel Core i9-10980 83.64% 88.18% Cascade 14nm 18/36 3.0 / 4.8 GHz 165W Threadripper 3960X 78.03% 82.32% Zen 2 24/48 3.8 / 4.5 GHz 280W AMD Ryzen 9 3950X 77.82% 84.25% Zen 2 16/32 3.5 / 4.7 GHz 105W Core i5-10600K 77.82% 86.89% Comet 14nm 6/12 4.1 / 4.8 GHz 125W AMD Ryzen 9 3900 77.64% 85.29% Zen 2 12/24 3.8 / 4.7 GHz 105W Ryzen 7 3800XT 77.60% 86.12% Zen 2 8/16 3.9 / 4.7 GHz 105W Ryzen 7 3800X 77.08% 85.50% Zen 2 8/16 3.9 / 4.5 GHz 105W Ryzen 5 3600XT 76.31% 84.97% Zen 2 6/12 3.8 / 4.4 GHz 95W Ryzen 5 3600X 73.91% 82.24% Zen 2 6/12 3.8 / 4.4 GHz 95W Ryzen 7 2700X 63.98% 71.64% Zen+ 8/16 3.7 / 4.3 GHz 105W Ryzen 5 2600X 61.99% 71.38% Zen+ 6/12 3.6 / 4.2 GHz 95W Ryzen 7 1800X 58.21% 65.19% Zen 8/16 3.6 / 4.0 GHz 95W Ryzen 5 1600X 53.93% 62.13% Zen 6/12 3.6 / 4.0 GHz 95W |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/11/20 at 00:40:10 https://liliputing.com/2020/12/microsoft-brings-x64-emulation-to-windows-on-arm.html Microsoft brings x64 emulation to Windows on ARM Windows on ARM was supposed to deliver the best of the both worlds: the low power consumption, long battery life, and always-connected capabilities that come with ARM-based chips, and compatibility with millions of Windows applications. In practice, it’s been more of a mixed bag. Not all Windows applications have been compatible, and performance hasn’t always been stellar. But at least Microsoft is tackling one of those issues – the company has announced that the latest preview build of Windows 10 on ARM adds support for x64 emulation, which means that 64-bit apps designed for computers with Intel and AMD chips can now run on PCs like the Surface Pro X, which have Qualcomm Snapdragon processors. As the ARM PC chipsets roll out this year and next, being able to run your mass of Microsoft stuff becomes relatively more important to various folks (those that will not run Linux for example). Once again MS will charge you LOTS of money to do this, but you knew that little factoid already. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/11/20 at 22:17:49 https://www.thestreet.com/investing/amd-blockbuster-deal-to-acquire-xilinx https://www.datacenterdynamics.com/en/news/ibm-and-amd-team-confidential-computing-ai-and-hpc-cloud/ https://www.nextplatform.com/2020/01/14/the-inevitability-of-fpgas-in-the-datacenter/ Computing is going to be hitting several hard tech walls inside the next 5 years. Going forward will require NEW STANDARDS to be written for the use AI and various accelerators of all sorts since big CPU node shrinks will no longer be available to drive progress. 5nm is rolling out now, 3 nm rolls out in two years and 1.5nm rolls in four years. Software developers really need something KNOWN and well documented to write their upgrades against, and Intel is showing everybody the danger of a separate company rolling their own "secret standards" in this arena. IBM has written more cooperative industry standards than anybody over the years, as well as doing more basic tech research than any other company. IBM understands RISC computing and mainframe, having built that stuff from the concept forward for decades now. Xilinx understands accelerators and AI and FPGAs better than anybody having invented it. AMD knows CPU tech and graphics tech and firmly believes in developing standards by working with the best in the industry to write those general standards. While Intel fumbles in a dark closet off on their own, look to see AMD, Xilinx and IBM begin to pull an industry standards group together. Realize that real growth in this industry come from the data centers, not from desktop PCs. Strategic thinking has to group around the most profitable side of your businesses, not the flashiest side of things. IS THIS A GOOD BUSINESS MOVE FOR THE VARIOUS PARTNERS? yes |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/12/20 at 06:55:52 Reality Check: I got a notification I had a two kernel upgrades waiting in que, the original and the major fixes reissuance due to issues that were found during original roll out. Ths was stated as two full kernel upgrades after it went through Ubuntu and then the Mint adoption processes. Every support package was listed twice as they all got upgraded twice (building one on the other). Both my browsers were there, with Firefox getting triple pumped due to its own internal changes as the Linux kernel related items only got double pumped. All my graphics this and that got at least one upgrade. Ditto for the Libre Office and other major softwares. Since I have been on Linux Mint, this is the biggest single update package I have ever seen come over the wire, ever. I noted the time and started it installing. Including the one hard reboot at the end, the entire thing took 6 minutes to do, complete. Amazing ..... Folks, in the days of 2400 baud modems it took me half a day just to download all the raw software. My main C: drive SSD really does make the largest difference in update speeds, as it can read and write the changes about as fast as the screen can keep up with the stuff flowing through the high speed cable modem. Six Minutes to install a really major update ......... only one (1) hard reboot at the end. Modern Linux Mint, ya gotta love it. http://https://static.greatbigcanvas.com/images/singlecanvas_thick_none/getty-images/roan-mountain-stream-in-fall,1942588.jpg Let's look forward a bit to when 4 threads per core SMT becomes real, as you know it will happen on Linux first. Assuming (a big ASSume here on this one) that my old Core DUO will be supported on the first wave, I will suddenly be "thread count equivalent" to just about all the windows running Intel processors from the last 15 years or so. Not bad for my old aging dual core rag-a-muffin Dell box. Now, looking forward to the future there is a Linux software shift coming right at you ...... lots of older packages like Flash are getting dropped by the distros and the general recommendation is now to do a separate media full install (usb preferably, DVD if you must) on your next update. Many of you who have Timeshift installed and working have noticed that your boot media (hard drive) is getting sorta full ---- this is because you are backing up some bigger and bigger over the air updates as the system sees them as changes that must be backed up. Plus, your old stuff and old games and all the other detritus you have built up over the years are bulging your waistline more and more ....... As I look at my stuff, I see things that I might not even bother reinstalling as I seldom if ever use them. So a general housecleaning isn't a bad thing for me. In any case, Linux Mint is forcing you to do a clean install between 19.3 and 20.0 because the changeover in operating system modes itself is causing some minor breakage. Silly stuff, like wifi breaking repeatedly ....... still, no point in going there as a clean install off of media avoids all these problems. =================================================== WARNING !!!! Ubuntu has made a right mess of this 20.04 LTS main distribution. Stay on Mint 19.3 until late next year at the very earliest --- do not do the 20.04 upgrade right now. Clem apologizes for the huge rat's nest Ubuntu has gifted him with. This clusterfook will likely cause some folks to turn away to other distributions as Mint can't fix it right now because it is still broken inside Ubuntu. And Ubuntu is foundering around, regressing this and that and basically messing it up weekly (differently each time) making it even worse and worse. You get current instructions to go to menu>system>passwords to fix an item ....... so you click menu>system>passwords --- and it isn't there any more !!! It is somewhere else now !!! Periodically, Mint discusses leaving Ubuntu as an operating basis and one time that I can remember it got serious enough for Clem and group to actually roll out a Mint variant based on Debian Rolling Stable (the stuff Ubuntu is based upon) ...... and we may be there again very soon if Ubuntu cannot get their act together any better than they are doing right now. I repeat ..... DO NOTHING ..... DO NOT UPGRADE TO MINT 20.04 AT THIS TIME. We are promised that 20.1 will get better and that most of the way through the next 20.2 period there will be a PAINLESS automatic upgrade path from Mint 19.3 to 20.2 provided by the Mint folks. Yep, it will take that long for Ubuntu to quit wiggling all over the place. Having been through the current painful fix it yourself method, I recommend WAITING until Clem's automatic method is all proven out. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/15/20 at 04:22:53 Having gone through a rough patch with a bad Ubuntu upgrade flowing up through Mint 20.04, I now firmly realize several things. All my separate programs are really just two items, the LibreOffice Suite (5 products) and Steam. This old separate program stuff comes in two chunks that are discrete and each run off their own set of rules. Both reside inside Mint's Software Manager, who takes care of installing them and keeping them up to date. As such, I can rape and scrape my old Dell's hard drive and put it all back inside an hour. I once had a much larger mass of older programs and older functions that I simply don't use any more. They have all been replaced by net based stuff that is all listed in Chrome now (and I get my Chrome stuff back now instantly by logging into my Gmail account. SuzukiSavage.com is simply one example of this general shift to the web. The items I consider to be CRUCIAL or KEY ITEMS all come through my Gmail backed up Chrome browser bookmarks and these backed up sites and preferences are all stored for me by Chrome. Just like my phone, all my backed up stuff is recovered through my Gmail name and password. I really don't use MS products any more, having been jacked around by MS more than enough until I simply don't have any of it any more. And I don't miss MS any at all either. Wife still uses it, and she gets tired of hearing me talk about her MS being a pain in the butt when it has to be fixed. (really, it is a very expensive pain in the butt too) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/15/20 at 18:49:04 https://liliputing.com/2020/12/lilbits-google-acquires-company-that-lets-you-turn-windows-pcs-into-chromebooks.html Google acquires company that lets you turn Windows PCs into Chromebooks http://https://liliputing.com/wp-content/uploads/2020/12/cloudready.jpg A company called Neverware has been offering software called CloudReady that lets you turn old computers into pseudo Chromebooks since 2015. But in order to do that, the company has been using software based on Chromium OS, the open source version of Chrome OS. Now Google has acquired the company and, among other things, that means that CloudReady will eventually become an official Chrome OS release with more of the features you’d get if you bought an off-the-shelf Chromebook. When that happens, existing users will be able to upgrade automatically, while new users will get the latest software. In the meantime, Neverware says there will be no changes to its pricing or support services — there’s a free version of CloudReady for home users, as well as Education and Enterprise editions available for $20 or $49 per year per user. Soon you will have another good way to continue to use your old Windows hardware instead of just trashing it out ....... ;) It will be interesting to see how Google handles the "end of life" obsolescence issue on this stuff since a good Linux distro could keep the same hardware running just about forever. Cloudready has been used in school systems for 6 years now to re-purpose out of date Wintel computer labs and "to extend the use life" of some out of support Chromebooks that still had sturdy hardware. You stick a button USB drive in a free port and set the machine's BIOS to boot from USB so the kiddies get their "Chromebook" started automatically. School systems are not made out of money and tricks like this are appreciated by supporting teachers, IT staff, parents, etc. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/18/20 at 19:32:03 https://liliputing.com/2020/12/report-microsoft-designing-custom-arm-chips-for-servers-and-maybe-surface-devices.html http://https://liliputing.com/wp-content/uploads/2020/12/surface-pro-x-700x467.jpg Rumor Report: Microsoft designing custom ARM chips for servers (and maybe Surface devices) Bloomberg reports that Microsoft is designing its own ARM-based chips for servers, and “exploring using another chip that would power some” of the company’s consumer-oriented Surface computers. The move wouldn’t exactly be coming out of left field. Microsoft already has a version of Windows 10 designed to run on computers with ARM processors, and the company already sells Surface Pro X tablets with Microsoft SQ1 and SQ2 processors… but those chips are basically tweaked versions of Qualcomm’s Snapdragon 8cx gen 1 and gen 2 chips. Designing its own processors would give Microsoft the ability to optimize hardware and software to work together to offer better performance, much the way Apple has been able to do with its iPhone, iPad, and now Mac processors. Rumor has it that Google, (which already designs its own chips for servers) may also be planning to develop custom chips for its own smartphones and Chromebooks. WHEEEEEEE now, doggie....... Look at all the people WHO DO NOT LIKE THE IDEA OF NVIDIA CONTROLLING ALL OF ARM's TECHNOLOGY, some big hitter folks who are now starting to roll their own ARM based chip designs. I bet some of these big hitter folks will use some RISC-V technology or some TENSOR technology in their chipsets for that exact same reason. Next, the head hunters will be calling on the better ARM software engineers with some hot job offers from the big boys ....... Lastly, what do you think this sort of environment means for poor old Intel? As ARM and RISC-V gain market share both Intel and AMD lose those potential sales units. Note I now speak of "x86 sales being hurt" in general --- not the little "shift some from Intel to AMD" effects that we have been watching over the last few years. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/20/20 at 16:13:42 HONEST, uncheated benchmarks between Apple, AMD and cheating liar Intel https://www.pcworld.com/article/3600897/tested-how-apples-m1-chip-performs-against-intel-11th-gen-and-amd-ryzen-4000.html?page=2 This is the second page of the PCWorld article (it contains all the graphs). Article is long to read and is complex, but the graphs show the AMD all-in-one laptops skunking Tiger Lake from Intel while slightly beating the Apple M1. Mind you, not on all tests ---- if you want Intel style "pure gaming" results look for single thread (short duration) performance graphs and Tiger Lake from Intel does better, but at much greater power draw current and much higher and faster heat build up numbers. If you game for long (for over 4 minutes) the Intel advantage completely evaporates due to heat build up and extreme CPU and GPU throttling, so you have to ask yourself how real Intel marketing's "claimed advantages" really are. ....... i.e. not very real Comparing the two x86 contenders against Apple M1 laptops show that M1 is a contender when using the new Cinebench R23 test that was recompiled to run native to Apple M1 standards, but this is still not a general across the board winner --- that role goes to the AMD Ryzen 4700 and 4800 APU units with 8 cores and 16 threads. Cinebench R23 Performance Maxon released a new version of Cinebench R23 with native M1 support, but there are some other key changes which should be mentioned as well. With Cinebench R20, the benchmark would run render a single scene and produce a score based on its completion. With Cinebench R23, the render scene is the same, but Maxon has interestingly changed it to render the same scene over and over for 10 minutes ALLOWING THE PROCESSORS AND LAPTOP THERMAL SYSTEMS TO WARM UP TO THEIR "NORMAL RUN" DEFAULT THROTTLING POINTS. The benchmark will even run beyond 10 minutes if the scene is still rendering when the timer hits zero. This longer test run acts to prevent some of Intel's little Intel Marketing games that took place with the older versions of Cinebench testing metrics. Larger numbers are better on this test. If you don't see the numbers, you are not able to see the whole bar length. (hit your scroll bar below the post, or right click on the chart to "open image in a new tab" to see all of it in the same glance) http://https://images.idgesg.net/images/article/2020/12/m1_cinebench_r23_nt-100870772-large.jpg On a six-core Intel H-class chip, Cinebench R15 takes 34 seconds to complete while Cinebench R20 takes about 108 seconds. With Cinebench R23 it now takes a minimum of 600 seconds to run under the new method. On a desktop or workstation with far more cooling it’s not an issue, but on laptops an all-core test that runs for at least 10 minutes can be far harsher—especially on CPUs that either make more heat, or laptops with more limited cooling. It’s actually called a “throttle test” which is a different method than before for Cinebench. None of this bothers Apple’s M1 much though. Based on TSMC’s most advanced 5nm process, it’s a stone cold killer, with Macworld reporting no fan noise at all during the run. That can’t be said of the x86 laptops, which all vary from fairly quiet to a little rackety. The performance is impressive though, with the eight-core M1 Mac now ahead of the four-core 11th-gen Tiger Lake as well as the older six-core Core i7-10710U. But if you give Intel’s older six-core more thermal head room it’s almost dead even with the silent M1. AMD, however, just shrugs at all this and yawns as all of its eight-core Ryzen 4700U and the various Ryzen 4800H series chips easily beat all comers when set to their performance modes and after being run for a few minutes. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/22/20 at 01:34:02 https://www.game-debate.com/amp/news/29310/newsAmpPage.html Intel is getting sued for misleading investors on 7nm timetable Written by Chad Norton on 2020-08-02 Last week we heard news of Intel delaying their 7nm chips by 6 months until late 2022 or early 2023, this resulted in a major crash of Intel’s share prices (and allowing AMD to pull ahead for the first time ever). Now, a famed US law firm is suing Intel on behalf of investors as they accuse Intel of not sharing the delay information earlier to investors. The Hagens Berman law firm is now calling for investors to join a class-action lawsuit against Intel for investor’s fraud. The official statement reports: “The investigation centers on whether Intel misrepresented and concealed manufacturing and performance issues with its next generation 7-nanometer chips.” Apparently, in 2019 Intel continuously claimed that their first 7-nanometer chips would start shipping in 2021, which was well received considering the continued delays to the 10nm chips and the reported doubling of area efficiency from 10nm as well. Intel repeatedly assured investors, once again reinstating the 7nm timetable. But of course, plans changed after Intel supposedly found a defect in their 7nm chips, causing the recent delay. “We're focused on investors' losses and whether Intel misled investors about the 7nm schedule and related manufacturing issues,” said the Hagens Berman partner leading the investigation, Reed Kathrein. It is getting really bad when your own stockholders class action sue for you consistently lying to them in stockholder's reports. :P Should the United States DOE join this class action suit since Intel lied to them repeatedly in the same time frame and cost them all the mega bucks DOE sank into the Intel Aurora Supercomputer? Intel's false marketing of Tiger Lake is another item that is worth another class action lawsuit. (read post directly above and directly below this one) This is part and parcel of Intel's conscious choice of lying to people in order to get money from them. Intel sux ........ Bob Swan's mouth did all the lying so it is Bob Swan who should be in prison right now. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/22/20 at 23:38:15 https://liliputing.com/2020/12/kfc-and-cooler-master-introduce-the-kfconsole-gaming-pc-no-really.html KFC and Cooler Master introduce the KFConsole gaming PC (no, really) https://youtu.be/73SqN-ueP7g Yes, click on the YouTube and see this commercial take on exactly what you can do with all that excess Intel Tiger Lake heat. Finger Lickn' Good, jest let Intel Tiger Lake heat that chicken up for your dining pleasure ....... ::) http://https://liliputing.com/wp-content/uploads/2020/12/kfconsole_04-150x97.jpg http://https://liliputing.com/wp-content/uploads/2020/12/kfconsole_05-150x138.jpg ..... will next year's hotter running more powerful 10nm Rocket Lake classify as an air fryer for the french fries ???? |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/24/20 at 13:58:40 https://liliputing.com/2020/12/lilbits-mediateks-rise-an-e-ink-monitor-and-marrying-arduino-and-raspberry-pi.html Look Ma, no Huawei anywhere ...... http://https://liliputing.com/wp-content/uploads/2020/12/counterpoint-smartphone-chip-share-700x477.jpg With a product lineup ranging from entry-level to premium chips, Qualcomm has dominated that space in recent years. But according to a new report from Counterpoint, MediaTek took the top spot for the first time in the most recent quarter. MediaTek also has a wide range of solutions, but the company is probably better known for its mid-range chips than for its high-end solutions. And while flagship phones grab all the headlines, the truth is an awful lot of mid-range phones are sold, which certainly helped with MediaTek’s rise. Counterpoint suggests the US trade ban on Huawei probably didn’t hurt either. Just as Intel has lost all of its crowns in PC land, Qualcomm has lost its leadership in Phone land. Again. Twice, actually. Once to Huawei (who designed their own and built them at TSMC ---- until Huawei was totally cut off by Trump's trade war) and now this time by MediaTek who designs bog standard ARM stuff and builds them at TSMC. The oriental market does not award "best" any longer unless you are Apple, the rest of the market always goes to the best price while swinging good enough features. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/24/20 at 23:32:44 avx512 What is it? Is it really a 14nm version of tech that should have been 10nm or 7nm when it was designed? Does it really include an Intel only (biased) AI accelerator function ??? a wee small voice from 4 weeks out in the future comes back to say "Yes, it does." Intel's new Intel Marketing buzzword for a "programming standard" that matches a new "Intel only wider/faster throughput bandwidth system" that isn't based upon any sort of Industry Standard, nor is it as good as what AMD chipsets have already been using (which IS based upon an old very wide mainframe bandwidth standard that has a complete existing fully documented set of written rules already out there) but avx512's hardware is simply isn't as good as the old mainframe stuff, just like the XE Graphics that Intel can build to right now isn't as good (considering the lithography constraints Intel is working against). Intel cannot compete against the world any longer, so they hide behind a "custom standards" smokescreen and CHEAT and LIE for all they are worth. So Intel now pre-hypes avx512 as "the greatest and very best thing ever" to try to get OS and software people to pre-adopt it. Same thing as Tiger Lake ---- a new sea of Intel Marketing lies floating a brand new turd in the punch bowl from Intel. Intel BS Marketing at its finest ..... avx512 is not available to be tested, but go ahead and believe us when we say it is really great stuff. Rocket Lake will use this "Intel only" smokescreen. Rocket Lake is NOT up to par thermally nor performance-wise against Apple nor AMD nor the newest generation of ARM Computer CPUs using the 5nm ARM standard designs (example Qualcomm and Microsoft). Intel Rocket Lake uses several tricks to appear better than it really is --- avx512 and the old Intel magic minute throttling tricks plus a dollop of hidden AI code written against a set of pre-customized benchmarks that come to the reviewers in an Intel supplied loaner box which will embody all of Intel's dubious progress for Rocket Lake (which is coming supposedly very early next year). Do you think the computer press has learned nothing from the funky Tiger Lake BS introduction? We shall see ........ ::) Intel, Cinebench 23 is jest a waiting for you as a real non-biased benchmarking standard when you finally do ship some real units, you little tricky poo, you. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/27/20 at 21:10:53 https://www.pcgamer.com/rocket-lake-proves-intel-is-broken/ Intel's manufacturing is broken and the newly announced 14nm Rocket Lake CPUs prove it By Jeremy Laird November 18, 2020 A sudden gust of wind blows the edges of Intel smoke screen off the new 14nm Intel Rocket Lake. The next six months are make or break time as Intel hurtles towards existential crisis. Tales of Intel’s woes have become so routine of late that the very existence of Rocket Lake and what it implies has been largely overlooked. Rocket Lake says very, very bad things about the viability of Intel’s entire business model. And that, in turn, makes it very significant for the PC as a whole. All of a sudden, Rocket Lake is now a 14nm backport, one that was never planned nor designed that way previously. Rocket Lake, of course, was Intel’s next 10nm desktop CPU architecture. As of this week Rocket Lake is suddenly a 14nm backport of Intel’s 10nm Sunny Cove CPU core architecture, as seen in 10th Gen Ice Lake notebook chips. Rocket Lake won’t be released until next year, which means Intel will be launching a new CPU design in 2021 on the ancient 14nm node. (Intel’s original plan was to move to 10nm in 2016) Yes, really. 14nm. Again 14nm ...... Intel has tried to sell the whole ‘back porting’ thing as a positive, a sort of groovy and inclusive approach to CPU manufacturing. “Hey guys, relax. We’re flexible, we can port from node to node. It’s freestyle. It’s all good,” Intel seems to be saying. The reality is that there’s really no such thing as a node-agnostic CPU architecture. It’s going to cost a huge amount of money to port those 10nm Sunny Cove cores, PCI Express 4.0 I/O and Xe-based graphics to 14nm for Rocket Lake. The die sizes will have to expand, the core counts will have to go down and it is questionable if XE-based graphics are going to work right on a back ported 14nm lithography. So, it’s not groovy or flexible. It’s a move made out of desperation because Intel’s 10nm production node still isn’t good enough for the prime time. Let’s repeat that. Rocket Lake will be launched in 2021 at 14nm because Intel’s 10nm still won’t be good enough for a desktop CPU launch. If this is true, then you can count on Intel HAVING to pay TSMC to build their processors for them in 2022-23, because Intel will be 3-4 full lithography nodes behind the competition by that point in time. :P 14nm to 10nm to 7nm to 5nm with 3nm being in trial runs at Apple and AMD by the time 14nm Intel Rocket Lake ships in any real volume --- I would count 4-5 full lithography nodes behind at that point, personally. Next, 14nm Rocket Lake will be just as physically big as an old style Intel processor --- Rocket Lake may actually have to lose a core or two compared to its original press releases for these now recognized size and thermal overheating reasons. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/28/20 at 22:11:24 https://www.extremetech.com/computing/318065-leaked-benchmarks-paint-conflicting-picture-of-intels-rocket-lake Leaked Benchmarks Paint a Confused and Conflicting Picture of Intel’s Rocket Lake There’s been a flurry of leaks around Intel’s upcoming Rocket Lake desktop CPUs, and it points in opposite directions as far as what to expect from the CPU. Most recently, there are some SiSoft Sandra results via THG that show an 8C/16T Rocket Lake chip running below the performance of a Ryzen 3 3300X with a claimed 1.8GHz base clock and a 4.4GHz boost clock. Last week, data from Ashes of the Singularity and Geekbench implied that the new CPU was an approximate match for the Ryzen 7 5800X. Both can’t be true, but the higher, better figures are more likely to be Rocket Lake uses Intel’s Cypress Cove 10nm architecture on a 14nm process. Cypress Cove is basically Intel’s Sunny Cove (Ice Lake) CPU core, back ported to 14nm. Sunny Cove, if you recall, was a genuine IPC uplift for Intel, but it traded back almost as much frequency as it gained in IPC. As a result, CPU performance in many applications was flat on a core-for-core or clock-for-clock basis. The problems that forced lower clock speeds on Intel’s 10nm node with Ice Lake, however, is said by Intel to not exist on the older 14nm node that is used here — and that, in turn, suggests that Intel should be able to deliver a net CPU performance improvement with this back port shift. At the same time, higher IPC typically requires more power consumption at a given frequency, and Intel cut the total number of CPU cores on this chip to eight, down from 10 because of size and energy cost (overheating ??) concerns. We should also expect to see signs of higher per-core power consumption causing deeper/earlier motherboard throttling on Cinebench 23 type testing. One has to question why Rocket Lake's top of the line chipset is compared to an older out of production Ryzen 7 instead of a modern top of the line Ryzen 9? Or at least to a CURRENT PRODUCTION modern Ryzen 7X class chipset? The answer is that Intel only advertises against something they can say they beat in some fashion ...... http://https://www.extremetech.com/wp-content/uploads/2020/12/GeekBench-Leaks.png ::) Note this chart says that AMD is still ahead of Intel's Rocket Lake right now and there is yet a whole new wave of AMD due to be out before this match up actually takes place in finished units that will be out next year. As with Tiger Lake, all we know about Rocket Lake comes from Intel's PR Machine. This is a very dubious source for any form of benchmarking information, as all of the PC industry is learning more and more to their dismay. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/29/20 at 00:20:37 https://www.tomshardware.com/news/intel-11th-generation-rocket-lake-cpu-sisoftware-benchmarks Unnamed Intel Rocket Lake Eight-Core CPU Fails To Defeat Ryzen 3 3300X which only has 4 cores http://https://cdn.mos.cms.futurecdn.net/26gXPJfRSM6TV3NmsoWpvL-970-80.jpg.we A fresh round of Rocket Lake benchmarks (via Tum_Apisak) has set the rumor mill into action once again. The highest Rocket Lake configuration so far has been eight cores, so this nameless 11th Generation processor may be the Rocket Lake flagship chip or a variant of it. If you haven't been following Intel's products, Rocket Lake marks the end of the Skylake era. Exercising Intel's latest Cypress Cove cores, Rocket Lake was poised to bring a real performance upgrade for users, unlike the previous iterations of Skylake that only offered minor improvements to existing architecture to justify the introduction of a new product. Intel has deferred to say how much performance we should see exactly, but boasted about IPC (instruction per cycle) enhancements in the double figures. SiSoftware listed this particular Rocket Lake-S sample with an eight-core, 16-thread configuration with 16MB of L3 cache. This description certainly matches that of the core specifications of a presumed octa-core Core i9 SKU. The Rocket Lake part reportedly runs with a 1.8 GHz base clock and 4.4 GHz boost clock. The processor's clock speeds are pretty far away from the values that we previously saw for another unidentified Rocket Lake processor, which flaunted a 3.41 GHz base clock and 4.98 GHz boost clock. In all likelihood, the SiSoftware chip could be an early engineering sample with a low 65W or 35W TDP. On the flipside, SiSoftware may be misinterpreting the clock speeds, which happens a lot when it comes to unreleased hardware. Intel 11th Generation Rocket Lake Benchmarks Rocket Lake (8 Cores) 431.10 Mpix/s 7.55 GB/s 28.32 kOPTS 21.79 GFLOPS 325.18 Mpix/s 4.04 kPT Ryzen 3 3300X 474.62 Mpix/s 13.59 GB/s 42.92 kOPTS 28.08 GFLOPS 274.02 Mpix/s 5.32 kPT Core i5-9600K 473.90 Mpix/s 10.85 GB/s 44.52 kOPTS 30.50 GFLOPS 354.65 Mpix/s 4.16 kPT Without a model name to pin the 8 core Rocket Lake processor to, it's pretty hard to deduct which processor it's replacing. Furthermore, we don't have all the details of the platform that was used in the tests, so we suggest you take these results with a pinch of salt for now. The results from the SiSoftware submissions aren't very inspiring though. The octa-core Rocket Lake processor seemingly put out worse scores than the Ryzen 3 3300X quad-core processor and Core i5-9600K hexa-core chip. Not performing as well as the existing Intel Core i5 processors and being clearly inferior to Ryzen 3 has been confirmed now in another set of samples reported by a second big name review house. This sort of results could mean "delayed due to something or another" and then a quick death for the Rocket Lake line of products ....... Intel will either fix it or kill it ...... fast ....... Bob Swan will not spend any more money on this one, he is too good of a bean picker to make that mistake. AI and some benchmarking & clever marketing lies clustered around a trick filled product introduction won't fix this one like it did for Tiger Lake as this the bloody wound is bleeding too fast and is simply too too big and ugly to slap a bandage on it. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/29/20 at 10:53:04 Last week and again yesterday a large chunk of FUD hit Intel's Rocket Lake introduction. Today an enthusiastic wave of exaggerated counteractive claims for the Rocket Lake chipsets came rolling up the beach to splatter against the large blocks of FUD left over from last week. Neither set of stuff should be taken to the bank. The FUD chunks were posted by some of the best big name computer review people, like Tom's Hardware and PC Gamer WHO POSTED THE REASONS for what they are concerned about the FUD items. Today's counter waves of exaggerated Rocket Lake optimism lacks the support data and the reputation of the FUD folks from last week. Use of hidden AI functions is suspected here (and has now been verified by independent testers who haven't got access to the magic Intel AI sauce and can't get anywhere near the AI based reported numbers) The truth is likely in the middle, out there somewhere, as always. Intel shills vs PC press stalwarts, we have seen this act before, haven't we? We now go into the dark, muddy, murky turbulent 2 month period right before the initial "released" Rocket Lake chipsets are shipped out and builders actually build up some actual units for sale and the owners can test them and report the results in bulk. Intel is also known for seeding cherry picked hyper performing chipsets with certain pet reviewers to get Intel some "free advertising" that does not spill directly back on Intel for being found out to be just about pure BS (as seen in retrospect). Marketing-wise, Rocket Lake HAS to beat the current Ryzen X series from AMD --- this is a given. Otherwise, why build them? ::) Also note that Lisa Su at AMD is holding back the last entire Zen 4 design generation of AMD processors in total silence until Intel finishes rolling out Rocket Lake and the smoke settles. She has another Zen 5 generation after that (with a 5nm lithography shift in it) that is in manufacturing development. Why hold it back until Intel is done lying all over the place? All the better to crush you with, my dear ...... crush you repeatedly if that is necessary as AMD runs to a 6 months cadence and the AMD process time at TSMC is booked a year in advance so the cadence is fixed for the next several years. AMD simply isn't responding to Intel's AI fiddles and such. Lisa Su and Apple just make sure that good benchmarks like Cinebench 23 get released and are out there in a timely fashion. This is all they need to do to burst the fraudulent Intel soap bubbles. The ARM PC chip boys are coming out next year, and all them hockey stick boys are going to be flying all over the place jest a slashing and bashing each other with the newest ARM Computer chip designs (some of which are quite capable, BTW). Intel isn't going to do well in the middle of that hockey stick Piranha feeding frenzy. Lisa Su is holding back on that last zen 5 AMD generation for countering the hockey stick guys as well. =================================================== Here is a rational voice rebutting some of the very recent frenzy of pro-Rocket Lake "mis-information". Intel Marketing is indeed flogging the AES-XTS test which does benefit from hidden AVX512 AI acceleration for the majority of the Rocket Lake single core improvement numbers. Intentional hyperclocking counts for a lot of these improvement numbers, but the resulting overheating is going to have to be dealt with as Cinebench 23 catches all that sort of action now-a-days. posted by dragontamer5788 I don't think AVX512 helps too much in many Geekbench workloads (maybe the AES workload but a lot of other stuff is INT64 based). But I'll have to double-check to be sure... Rocket Lake has bigger L1 cache... and also bigger L2 cache. It might also be a wider core, does anyone have the REAL information on the backend of the modified 14nm Cyprus Cove? The big performance jump is in indeed in the AES-XTS test which does benefit from AVX512 acceleration. The 10700K scores around 3.07GBps in the single-core test, this 11700K around 9.24GBps. There is very little difference in the multi-core AES-XTS test scores which means that the AVX512 reduces the multi-core clock quite significantly, to the point where it is hardly worth it. That one sub-test (the AES-XTS test which does benefit from AVX512 acceleration) on its own lifts the single-core score by 10 full percentage points, so the final real processor performance score is fairly misleading. Rocket Lake is an improvement, but not a 30% improvement. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/30/20 at 03:31:04 DO YOU CHOOSE TO DRINK THE HIDDEN ACCELERATOR KOOL-AID ??? AI and "hidden support softwares" are not present in all applications, so the AI stuff does not work all the time and it should not be given up to the public as "CPU" improvements in advertising or in benchmark test results. Class action suits are in motion even now for the Tiger Lake buyers who got defrauded by this Intel Marketing ploy. Where and how you spend your money is 100% how you vote in this issue. Apple controls their software universe, and if you need it they REQUIRE that it gets put into all software and drivers automatically. Apple also names their neat little tricks and states their use openly. AMD just builds faster and better processors ---- you sign up for your own tricks if you think you need any. Intel keeps secrets and lies to everybody, that is just the new Intel and how it rolls. You pay more for Intel so we hope the Intel Marketing tricks amuse you enough to give you recompense for the extra 30% more that the little Intel Inside logo costs you. Otherwise, go join a class action suit to get some of your money back. ==================================================== Just type this into your browser and pick the one that fits your circumstances best --- likely you can participate in more than one as Intel has provided multiple issues to sue them over ...... current intel class actions on the Google Chrome search engine https://www.google.com/search?q=current+intel+class+actions&oq=current+intel+class+action&aqs=chrome.1.69i57j33i22i29i30l5.12415j0j15&sourceid=chrome&ie=UTF-8 |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/30/20 at 18:10:20 https://www.anandtech.com/show/16205/intels-11th-gen-core-rocket-lake-detailed-ice-lake-core-with-xe-graphics Dr. Ian Cutress of Anandtech calmly reviews the latest Intel announcements on Rocket Lake With the new processors, Intel is targeting a raw instruction-per-clock uplift in the double digit range, which would be similar to the uplift we saw moving from Comet Lake to Intel’s Ice Lake mobile processors. Because of the node difference, the exact IPC change is likely to be lower than what we’ve seen before, but 10%+ is still highly respectable, especially if Intel is also able to maintain the high frequency it has achieved with the current generation of Comet Lake. http://https://images.anandtech.com/doci/16205/2_575px.jpg One of the benefits of moving to a back-ported Sunny Cove core will be the inclusion of the AVX-512 vector acceleration unit in Cypress Cove. This enables Intel to enable its library of Deep Learning Boost technologies for AI and ML acceleration, including support for Vector Neural Network Instructions (VNNI), finally bringing AVX-512 to the desktop platform. http://https://images.anandtech.com/doci/16205/3.jpg However, to mix and match the right combination of core count, graphics, and AVX-512 for die size/yield/cost, it appears that Rocket Lake-S will only offer a maximum of eight cores in its largest configuration. Within the press release PDF, Intel stated that the current silicon as tested is rated for 125 W TDP, with a top turbo boost of 250 W, which matches what we see on the Core i9-10900K already. There’s no escaping the performance-per-watt characteristics of the process node, which indicates that Intel might find hitting those high frequencies a little easier with fewer cores to deal with. Intel is also promoting new overclocking tools with Rocket Lake, however did not go into details. For those invited to the Intel event, AVX-512 was explained as an AI booster engine, which was given the moniker "Intel Deep Learning Boost". Intel is riding hard on AVX-512 to make up for the fairly underwhelming new processors they have had to build on Intel's existing 14nm processes. Irony is the AVX-512 boost tricks could be applied to the existing run of Core i5, i7 and i9 processors that Intel is building right now and still give somewhat similar benefits. This may happen in the future, making up a new, higher performing generation of AVX-512 boosted processors. Still, the massive Intel secrecy ends now ....... and that is a good thing. Look to see Intel start to REQUIRE water block cooling in the future as these things will need much much better cooling or all of their AI tricks and hyperclocking will throttle-wilt during Cinebench 23 testing. ================================================== AND tested they will be, empirically and against a set of representative tasks for a goodly period of time. As in Cinebench23 all processors will be tested for long enough to ensure they heat up and throttle down to their steady thermal state. Ergo water cooling will be becoming an important Intel requirement -- Intel wants the very best cooling system that they can get to enable the best performance that they can show. =================================================== Now all Intel has to do to become "real" again is to start talking about and advertising only SYSTEM PERFORMANCE NUMBERS and quit messing with words like "processor performance" or "CPU throughput numbers". Then the false advertising lawsuits will stop. =================================================== 5nm is looking like a very good, high yielding lithography node and I think money spent already by TSMC and others on 5nm will be well used in the future. Intel will never go there, Intel's most specious semi-real future plans seem to stop at 6nm (TSMC's long term bulk production standard) =================================================== All of the existing production system team pair ups will increase in importance as time wears on. Google and Tensor. Google and ARM. Apple and ARM, Qualcomm and ARM. Microsoft and ARM. NVIDIA and ARM, AMD and Xylinx, Intel and all its little minions (quite a list here with December 16, 2019. Intel Acquires Artificial Intelligence Chipmaker Habana Labs --- .April 16, 2019. Intel Acquires Omnitek, Strengthens FPGA Video and Vision Offering. ---December 28, 2015. Intel Completes Acquisition of Altera. What are we looking at? Folks should be planning for lithography downshifts to become rarer going past 5nm and be much less effective and start to take second place to improvements in AI systems integration. http://https://cloud.google.com/tpu/docs/images/tpu--sys-arch4.png Both Google and Nvidia have walked very far down this pathway and understand it very well ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/01/21 at 20:26:05 https://www.youtube.com/watch?v=iZBIeM2zE-I&t=198s Linus at Linus Tech Tips covers some of this and offers a few sets of numbers and comparative predictions about Rocket Lake vs AMD. He also asks that Lisa Su to PLEASE quit kicking the dead Intel donkey in the balls. Please stop kicking it, it is finally dead and it is not going to move again. :o ....... does this mean when the dead Intel donkey eventually appears to get back up it is actually a zombie donkey ????? http://https://vignette.wikia.nocookie.net/heroes-of-camelot/images/0/04/Zombie_Donkey.PNG/revision/latest/scale-to-width-down/333?cb=20140818162238 ================================================== For those who deem the effort to look worth to be worth the effort, the new Cinebench 23 has ranked all current processors by Multi-core results while still offering the best of the installed cores single best core speed numbers for reference. Note please, current modern games all use multi-core processing now so any BS about using that one single "best" core to rate chipsets has become just some more Intel Marketing BS. This INCLUDES the Apple M1 and will include all ARM chipsets when they show up, so it Cinebench 23 is a good reference benchmark because it can correctly rank the different brands and types of processors --- and it avoids many of Intel's magic minute, hyperclocking and "turbo boost" games as it does make sure the chipset warms up to its steady state and the processor is system BIOS throttled to its settled performance level. Cinebench 23 reviews entire systems including the cooling system (listing models and brands) as well as just giving out the averaged results for a given processor. Reminder, honest companies test completed production units and rank them using Cinebench 23 as this test impartially reports how good your entire thermal system really is. There is a lot to like about Cinebench 23 because it can sort out so much of the Intel Marketing BS easily and with no fuss ...... https://www.cgdirector.com/cinebench-r23-scores-updated-results/ (click to the left to see pretty graph of this data) CPU Name Cores Ghz Single Score Multi Score AMD Threadripper 3990X 64 2.9 1262 75671 AMD Epyc 7702P 64 2.0 993 48959 AMD Threadripper 3970X 32 3.7 1308 46874 AMD Threadripper 3960X 24 3.8 1307 34932 AMD Threadripper 2990WX 32 3.0 1005 29651 AMD Ryzen 9 5950X 16 3.4 1684 28782 Intel i9 9980XE 18 3.0 1114 27093 AMD Ryzen 9 3950X 16 3.5 1406 26375 Intel i9 10980XE 18 3.0 1063 25490 AMD Ryzen 9 5900X 12 3.7 1670 22046 AMD Threadripper 1950X 16 3.4 1027 19635 AMD Threadripper 2950X 16 3.5 1135 18797 AMD Ryzen 9 3900X 12 3.8 1312 18682 Intel i9 10900K 10 3.7 1415 18034 Intel i9 9960X 16 3.1 1075 17953 AMD Threadripper 1920X 12 3.5 1054 15038 AMD Ryzen 7 5800X 8 3.8 1596 14812 Intel i9 9920X 12 3.5 1067 14793 Intel i9 9900X 10 3.5 1182 13994 AMD Ryzen 7 3800X 8 3.9 1346 13848 Intel i7 10700K 8 3.8 1345 13302 Intel i9 9900K 8 3.6 1343 12470 AMD Ryzen 7 3700X 8 3.6 1345 12195 AMD Ryzen 5 5600X 6 3.7 1593 11201 AMD Ryzen 7 2700X 8 3.7 1102 10140 AMD Ryzen 5 3600X 6 3.8 1323 9526 Intel i7 9700K 8 3.6 1285 9428 AMD Ryzen 5 3600 6 3.6 1245 9073 AMD Threadripper 1900X 8 3.8 1005 8979 AMD Ryzen 5 2600X 6 3.6 1094 7523 AMD Ryzen 3 3300X 4 3.8 1299 6787 Intel i5 9600K 6 3.7 1196 6596 AMD Ryzen 3 3100 4 3.6 1105 5423 |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/05/21 at 08:24:21 https://www.techspot.com/community/topics/intel-core-i9-10885h-appears-slower-than-the-i7-10875h-by-almost-20-percent.265592/ Intel and Dell cannot keep their stories straight Intel seems to want to keep using Geekbench for everything as it best supports their various cute little hyperclocking and magic minute tricks and it allows Intel to hide their lack of progress under a weight of concealing numbers. This constant lying can lead builders to design inadequate power systems (not enough battery) and can result in thermal overloading of the cooling system (Intel had badly understated the power losses and resulting heat build ups on this one). This allows an otherwise good vendor like Dell to get sucked in and get stuck on Intel's BS, BS that is supported by all of Intel's resource materials leading a builder like Dell to get deeply embedded in a mess like this one, finally finding out from their end users that Intel lied too much and too deeply and for too long ........ and the resulting built by the thousands units are SLOW, throttle like heck and overheat. This past weekend the various Youtube reviewers were chuckling because Intel posted Dell information from several benchmarks that said that their i9-10885h-appears-slower-than-the-i7-10875h-by-almost-20-percent. Intel, it appears, has caught its own hand in the Dell cookie jar and this whole thing is becoming embarrassing for Dell ...... https://www.techspot.com/community/topics/intel-core-i9-10885h-appears-slower-than-the-i7-10875h-by-almost-20-percent.265592/ https://www.notebookcheck.net/Intel-Core-i9-10885H-is-almost-20-percent-slower-than-the-Core-i7-10875H.497644.0.html Oct 14, 2020 In a nutshell: The Intel Core i9-10885H has been found running 20 percent slower than expected in Dell Precision laptops, capping its clock speed at just 2.7GHz. This is most likely due to the limited cooling in this entry-level Dell device making this situation is a reminder to consumers and professionals to investigate performance in laptops and PC builds as a whole before investing in building bunches of them. Note: don't trust Intel, do your own testing using Cinebench23 and other thermal analysis tools before committing to production. Intel's vPro line of CPUs are business-class chips, typically aimed at IT professionals designed with onboard hardware security firmware. These features defend against attacks below the OS and allow companies to repair and manage their machines remotely. In a recent laptop benchmark comparison ran by NotebookCheck, the vPro-enabled Core i9-10885H has seen results nearly 20 percent slower than the Core i7-10875H, its consumer-oriented equivalent. With the same core count and cache, and boasting 200MHz higher boost speeds, the i9-10885H wins on paper. That said, several benchmarks on the laptop CPU as found in the Dell Precision 3551 showed roughly 20 percent less performance in multi-threaded tasks, and closer to a 10 percent deficit in benchmarks like 7-Zip, Blender, and LibreOffice. http://https://static.techspot.com/images2/news/bigimage/2020/10/2020-10-14-image-12.jpg At 100 percent utilization, the i9-10885H stabilized at just 2.7GHz, only 300 MHz above its base clock, so this 20 percent difference can most likely be attributed to Dell's cooling solution in the Precision 3551, causing bottlenecks in the CPU. This is an entry-level workstation laptop, coming in below the 5000 and 7000 series, and priced at around $2,100. Multiple reviewers have promised to Cinebench 23 the whole sticky mess to try to learn what exactly what the underlying truth may wind up being ....... (Dell is expected to only ship "corrected units" starting after January ---- this assurance assumes that there is just a settings error causing the whole mess instead of a total performance mismatch in setting up the CPU & cooling system of the laptop based upon Intel's published bad information) ::) class action lawsuit time again, folks |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/05/21 at 23:50:59 https://www.tomshardware.com/news/amd-patent-shows-CPU-FPGA-integration Stock based mergers can move very quickly as no funds are involved so government approvals are minimal. AMD and Xylinx are cooking up their first co-joined processors and are pre-announcing them through hard PATENT applications. AMD has been doing wonders in the CPU industry, with its well-received and high performing Zen 2 and Zen 3 based processors, but now it seems AMD wants to improve performance not just through faster cores, but through the use of FPGAs. Just a few days ago, AMD filed a patent for integrating FPGAs into a CPU, which would allow the processor to run custom instruction sets to extend the its capabilities. As a side note, this patent was made just a few months after AMD's acquisition of Xilinx, a company dedicated to making FPGAs. FPGAs, or Field Programmable Gate Arrays, are simple yet powerful devices that can run specific instruction sets very quickly. This is different from a standard x86 CPU core that's designed to run a near-infinite variety of instruction sequences, albeit sometimes slowly. If there's a specific task (graphics, physics, encryption, etc.) that's used regularly, it might be beneficial to create a custom instruction on an FPGA that will process the code much more quickly. Plus, FPGAs aren't limited to a single instruction set; they can be re-programmed to run another instruction set if necessary. This seems to be what AMD is going for, and AMD's implementation would allow the FPGA unit to share registers with the CPU itself. Simply put, this allows the CPU to very quickly offload instructions to the FPGA unit when necessary. We don't know what specific tasks AMD is looking at, but presumably anything currently using dedicated FPGAs or AI accelerators could see instant support. We also don't know where this FPGA (or FPGAs) would be located. If we're talking about a Zen 2 or Zen 3 based design, the FPGA could be installed on its own separate die (chiplet) connected via the infinity fabric. Alternatively, it could be integrated directly onto the CPU chiplet, sharing a die layer with the cores. This would be the most optimal setup as far as performance goes, but it would require new compute chiplets to be designed. Aiming for a 2-3 year implementation time frame, AMD intends to preempt anything Intel can come up with by having the ability to CHANGE their chiplets on the fly. Be afraid Intel, be very afraid ....... 8-) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/06/21 at 00:20:06 https://siliconangle.com/2021/01/03/report-huawei-planning-worlds-first-3nm-mobile-chipset/ Other breaking news, TSMC increases 5nm production capacity from 60,000 wafers a month to over 100,000 wafers a month (yields remain very high and the very quick 5nm ramp up is very encouraging). Apple is 100% in on 5nm and AMD is beginning their roll in to the new smaller technology. The Very First 3nm production chipset is being taped out for Huawei. Huawei feels that if they actually pilot develop a new technology they cannot be excluded from it by Washington DC "entity list" actions. The first 3nm chip is called the Kirin 9010 and very limited production will take place very early this year, in Q1 2021. All of the major players Huawei, Qualcomm and Apple have already placed TSMC production orders for 3nm sample production for this year. AMD has very large hard plans for 5nm this year and will be ready with a Xilinx chiplet layer integration effort by the time AMD 3nm rolls around in about a year and a half. 5nm already allows 14 layers on a chiplet, 3nm allows for 20 layers, but with a much larger "functional lay down area" per layer as the features are so much smaller. Everybody is hunting for their next future pathway planning for the lean times that will come after lithography shrinks become completely ineffective. New partnerships are being forged for that exact reason. After 3nm comes 1.5nm and the risk it won't actually work out is being acknowledged by TSMC and others. =================================================== TSMC has declined to build any extra capacity to do large scale work for Intel. Intel's designs are not easily adapted for TSMC's production flows and quite frankly, Intel seems at best more like a "possible" short term thing at this stage of TSMC's total production planning. 6nm Intel is not a business you build new facilities to cover. Plus Intel is known for having some really bad habits (habitually shitting all over their partners) so TSMC simply chooses to continue to see Intel as a competitor and not as a customer. Intel would only be welcome at TSMC for some single very discrete separate product, such as a GPU, and then only if TSMC built the entire thing for Intel and built all the development runs with all of it being "paid for in advance". Intel would own all the scrap and all the various design issues that came up ..... http://https://www.icinsights.com/files/images/bulletin20200213Fig01.png Intel ranks #10 in this list at ~ 850,000~ wafers (850 kw/m stated as shown in chart above) for all Intel sites per year. This goes to show just how small Intel has become on the world stage. Sounds like TSMC simply intends to let Intel go on down the toilet bowl swirl as there is nothing there really worth saving ....... Intel has reacted to this TSMC decision, cancelling all Intel internal production on their smaller capacity (or older technology) chipsets to make more room inside their current production envelope. Fab 42, the often cancelled upon Intel 7nm and 5nm facility is being partially converted to 14nm and 6nm formats ASAP, once again making more production room for what Intel can actually do for itself internally. Intel is actively circling their wagons now, getting ready for their last stand as the hockey stick boys have begun slamming their 6nm and 5nm hockey sticks against the ground in a rhythmic drumming while preparing to seriously attack Intel's market share with ARM CPU designs and start peeling off huge chunks of processor business. Intel has no friends anywhere any longer ...... even Microsoft is now a processor producing competitor to Intel. Intel is shedding market share so very quickly right now they will be able to fit inside Intel's current facilities quite handily. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/07/21 at 11:07:53 https://www.pcgamer.com/intels-11th-gen-desktop-cpus-could-be-split-between-two-different-architectures/ Buyer beware, rumor has it that not all of Intel's next-generation desktop CPUs will be based on the newly announced architecture --- yeah that one, the one that is supposed to deliver a double-digit gain in IPC (instructions per clock) performance and all sorts of other goodies. They will, however, all be confusingly labeled as 11th Gen Core parts. According to Expreview, however, Rocket Lake-S will be reserved for Intel's 11th Gen Core i9, Core i7, and Core i5 SKUs. Meanwhile, Core i3 SKUs will be a refresh of Comet Lake-S (as will upcoming Pentium and Celeron models), and will not support PCIe 4.0. Comet Lake Refresh CPUs will also lack Xe graphics. Instead, some models will be paired with Intel's UHD Graphics 630 solution with 24 execution units, while others will lean on the slower UHD Graphics 610 solution with 12 execution units. http://https://cdn.mos.cms.futurecdn.net/niNptq8QBm9kb7FsbVSUaR-970-80.jpg.we Smells like another BS game filled product introduction series from our buddies at Intel ........ Lots and lots of acetone wiped retreads of old processors being called "Gen 11" just to confuse ordinary folks who are buying PCs just by reading the boxes. "Hey Charlie, which one did you get, the good one or last year's retread with the slow processor and the last year's non-XE graphics ???" Hey, they never actually said ---- I definitely ordered the good one but now Intel says you can't test or benchmark their chipsets so I guess I'll never really know for sure what I got. I should've have ordered an AMD instead. The first ARM designed 8 core PC chipset shipped today, in China. This motherboard is running a Chinese Linux version as its only option. Rumor is that chip was produced in a state owned production facility using some end of life TSMC lithography equipment, and it was run at the original 7nm. Same set up has already produced and shipped a 24 core ARM design for the Chinese Military. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/10/21 at 04:01:06 We are 3 days away from broad AMD announcements by Lisa Su at CES 2021. What is AMD's state of supply and how does that affect what AMD announces for roll out plans? Last year AMD had to ramp their supply chain 2-3x in size just to cover all their brand new market share AMD had just taken away from Intel. Intel is shrinking, actually becoming a fairly good fit into their existing fab space. AMD is growing, adding in additional package assembly operations. Insiders are saying that AMD is building brand new buildings to house all the chiplet sorting and other functions for the 5nm chiplets that are rolling in soon from TSMC. AMD has lots of money now to spend on this effort, but simply getting it all done puts a functional brake on what AMD can announce in 3 days time. =================================================== A mental shift at AMD is taking place as AMD is now the PC processor industry leader instead of the ones playing catch up. The overall market for PC chipsets has expanded 25% just this past year and AMD has been repeatedly shocked at the amount of stuff flying off their shelves as it becomes the "preferred, best in class" chipset provider. We will watch the AMD and Intel presentations to see the reflections of this change. One change already noted is that Lisa Su will speak 2 days after Intel and Nvidia speak, letting Lisa Su gauge her announcements to balance out the competitive needs AND her supply line issues. =================================================== https://www.tomshardware.com/features/amd-vs-intel-cpus In the week leading up to CES Tom's Hardware rolls up the processor lists getting ready for all the new announcements. AMD vs Intel 2021: Who Makes the Best CPUs? http://https://cdn.mos.cms.futurecdn.net/ZT89RwnfySndiR8Favks9k-970-80.jpg This article covers the never-ending argument of AMD vs Intel desktop CPUs (we're not covering laptop or server chips) based on what you plan to do with your PC, pricing, performance, driver support, power consumption, and security, giving us a clear view of the state of the competition. We'll also discuss the lithographies and architectures that influence the moving goalposts. Overall, there's a clear winner, but which CPU brand you should buy depends mostly on what kind of features, price, and performance are important to you. You can see how all of these processors stack up in our CPU Benchmark Hierarchy, but the landscape has certainly changed in the wake of AMD's Ryzen 5000 launch. AMD's newest processors, the Ryzen 9 5950X and Ryzen 9 5900X, not to mention the Ryzen 5 5600X, have upset the entire mainstream desktop lineup. We've added in the new models and some commentary based on what we know about Zen 3 so far, but you can head to our expansive in-depth coverage of the Ryzen 5000 series, including pricing, benchmarks, and availability, for more info. Suffice it to say, the Ryzen 5000 series are the highest-performing chips on the market and beat Intel in every metric that matters, including gaming, application performance, power consumption, and thermals. Unfortunately, sweeping shortages driven by high demand have kept the Ryzen 5000 chips out of the hands of enthusiasts, and they've remained nearly impossible to buy. Of course, you could get lucky and score a Ryzen 5000 chip during the rare moments of availability, but if you must purchase a chip today, many of the other chips outlined in the article below will likely be your only option until AMD can begin to satisfy the huge demand for its chips. AMD vs Intel CPU Gaming Performance In the AMD vs Intel CPU battle, AMD holds the lead in the critical price bands, particularly right in the middle and high-end of its stack, but our benchmarks show the Intel's gaming performance is no slouch, either. Below we have a wide selection of collective gaming performance measurements for the existing chips in the different price bands. http://https://cdn.mos.cms.futurecdn.net/YqfpCwKBTtW3y5Xh2AKkw5-970-80.png.we AMD vs Intel Productivity and Content Creation Performance http://https://cdn.mos.cms.futurecdn.net/RRdzVfw6SuFJ4P5pqvevqg-970-80.png.we Tom's Hardware is blunt and accurate in stating that Intel simply isn't competitive any longer at any level --- then breaks it down with fine supporting details that will make Intel squirm in the dirt a bit. Tom's uses a combined benchmark called GEOMEAN which is composed of sT Lame, Cinebench, POV Ray, and Y-Cruncher each of which test the different areas that make up a processor. I find it interesting that AMD processors all come in closely grouped while Intel spreads itself over a span of results. We know that AMD processors all use the same chiplets, so that close grouping isn't totally surprising to me. So, AMD could announce ABSOLUTELY NOTHING in 3 days time and AMD would still dominate Intel at every level in the stack ....... but AMD has some strong progress to announce for 2021, yes they do. AMD could sell every chipset that is sold in 2021 with what they have right now if it was simply based on the merit of their processors ---- but the hard reality is that gross demand for new computers world-wide is at an all time high and AMD simply cannot build all of the processors that are needed world-wide. PC shipments are on a spiking trend, being up over 25% as reported by everyone who tracks PC shipments. Add in the laptop chipsets and you have even larger numbers of processors being moved right now. You can't physically buy one of the better more modern AMD chipsets, there aren't any to be had anywhere ....... Intel is out of some of their better models as well. Chromebook numbers are spiking strongly as well, as folks cannot get the PCs and laptops they really want so they settle for the Chromebook they can still get their hands on. Intel is shorting their supply of Chromebook chipsets to make more production room on their larger chipsets. AMD still sees their ability to allocate two of their low end chiplets to a Chromebook processor as a good use for their sorted out "bottom of the barrel" chiplets and AMD is actually increasing their overall Chromebook supply levels accordingly. ARM based Chromebook chipsets have started rolling in as well. Chromebooks are VERY HOT right now and it isn't even back to school time. Gartner says Chromebooks were up by 200-percent measured year-over-year. Several automotive companies have recently shut down due to "a global silicon shortage" that is also hitting the PC industry. This shortage is bound to somewhat delay generational improvements both at AMD and at Intel as both companies can sell all the chipsets that they can make with what they have right now. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/11/21 at 13:36:52 This week at CES Intel No specific facts, no specific announcements, no specific claims of industry leadership. Intel claims that they have the best most cohesive set of hardware and software to give the best solutions to business users. They imply they have the "best hardware performance", but actually will only say they have the "best most cohesive set of hardware and software". They never say how their stuff works to give you this"most cohesive software/hardware combination" --- but what they showed the computer analysts in private was some hidden AI speeding up a carefully selected set of benchmark tasks. So, the Intel's secret press presentation is just more Intel BS Marketing, more misleading smoke and mirrors that carefully dances short of any real easily caught lies. Intel is wary of the class action lawyers as Intel has lost a lot of money in class action legal actions last year and now have EU and USA regulators looking hard at their advertising practices. Remember, nothing that Intel "promises to show you" is one bit more than AMD can really actually do for you, for totally real, right now. https://liliputing.com/2021/01/intel-launches-11th-gen-core-h35-chips-for-gaming-laptops.html Brad Linder drew a few extra bits of information from the Intel Keynote address and even more from the private "invitation only" Intel show-and-tell afterwards. Intel says these new 11th-gen Core H35 series processors are 10nm processors designed for notebooks with discrete graphics, but they also included the company’s latest integrated graphics technology for up to twice the graphics performance of a 10th-gen Core H “Comet Lake” processor. The new chips also offer up to 15-percent better CPU performance than a Comet Lake-H chip, and more than 40-percent CPU performance than a 15-watt Tiger Lake-U processor like those that have powered most Intel-powered thin and light laptops to ship in recent months. So far there are 12 new members of the 11th-gen Core H family, topping out at the Core i7-11375H quad-core, eight-thread processor with support for single-core speeds up to 5 GHz. Intel has provided specs for a few of the new chips: Name Cores / Threads TDP Max Turbo 1-core / 2-core / 4-core Cache Memory Core i7-11375H 4 / 8 28 – 35W 5 GHz / 4.8 GHz / 4.3 GHz 12MB DDR4-3200 LPDDR4x-4266 Core i7-11370H 4 / 8 12-28W 4.8 GHz / 4.8 GHz / 4.3 GHz 12MB DDR4-3200 LPDDR4x-4266 Core i5-11300H 4 / 8 12-28W 4.4 GHz / 4.4 GHz / 4 GHz 8MB DDR4-3200 LPDDR4x-4266 But later this year Intel says it will bring desktop-class performance to an 11th-gen mobile H-series processor code-named Tiger Lake H. It’ll be an octa-core processor capable of hitting 5G Hz speeds on multiple cores simultaneously and it’ll support 20 lane PCIe for up to 40 Gbps throughput. What’s the difference between Tiger Lake-H35 and Tiger Lake-H? At least 10 watts of power. It’s likely that the chips launching later this year will run at up to 45 watts. (thermally suicidal high power draws for a standard laptop chassis) Other features for Tiger Lake-H35 include support for WiFi 6 and 6E, PCIe Gen 4 x4, and DDR4-3200 or LPDDR4x-4266 memory. ================================================== OK, it is a full day later and Intel has had to go back to its fans and EXPLAIN to everybody that VPRO was the big announcement that they made for CES this time around. VPRO means a very mild hardware bump (yes, we heard that part) that is "effectively magnified" by background software activities (yes, we heard about the AI working in the background to cheat on the benchmark performance of your chipset). But this is not CES level "progress", it is simply some more BS Intel Marketing tricks again. Intel, you SHOULD have multiple government efforts (USA and EU regulators at multiple levels) looking hard at you for defrauding your customer base if you have to go back to tell that customer base that some very mild BS Intel Marketing cheating was the sum of your progress for this year. "But later this year Intel says it will bring desktop-class performance to an 11th-gen mobile H-series processor code-named Tiger Lake H. It’ll be an octa-core processor capable of hitting 5G Hz speeds on multiple cores simultaneously and it’ll support 20 lane PCIe for up to 40 Gbps throughput." And this is a vague bogus sounding future roadmap claim, not reporting hard progress of any kind. You have been spewing these sorts of roadmap claims all along and they simply get replaced by more roadmap claims as you can never actually get around to doing them. So this sort of stuff from Intel gets routinely ignored by everybody now-a-days when you spew it at us. :P |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/12/21 at 07:00:55 https://liliputing.com/2021/01/samsung-exynos-2100-is-a-5nm-chip-with-arm-cortex-x1-and-support-for-5g-and-200mp-cameras.html Rumor time again ....... Qualcomm and Samsung's best phone chips feel very much like ARM X1 chipsets (lightly modified by Qualcomm and practically not modified at all by Samsung). Intel has placed their first production 6nm TSMC product placement order (a video card chipset). Mediatek is quietly growing larger, Mediatek is swinging a very strong AI boosted 4 core Chromebook chipset right now, with plans to push out a 8 and a 12 core ARM based Windows 10 laptop chipset later this year. These are not candidates for "best chipset awards" but are good enough to make it to consumers in some very cost effective laptop products. Mediatek is designing to use more and more RISC V sub-chips on their motherboards as they are much much cheaper and are still relatively plentiful right now. Mediatek already uses AI to speed up their products, learning all of Intel's predecessors tricks and simply doing it better than Intel does. Both of them learned the tricks from Huawei and Oppo who have cheated on benchmarks for years and years and years now ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/12/21 at 09:27:08 This week at CES AMD Microsoft, Hewlett Packard, Lucas Films, Industrial Light and Magic, Lenovo, Two Formula 1 racing teams, 5 different universities and medical companies from Science & Medicine all co-presented with Lisa Su. The conglomerated message is that the future of our lives is being pushed forward by a constant stream of higher capacity innovative AMD products. Contrast with Intel who would not even let CEO Bob Swan even show his ugly face (yes, he is a net negative right now and the rumor mill sees him gone sometime fairly soon). If Intel even had a partner to show out there, it wasn't apparent as all the main compute partners shunned Intel and actually presented for AMD this year. Lisa Su said Ryzen 5 processor designs are here now and brand new 5nm and 6nm lithography lines are coming for later this year and early of next year (just as soon as Apple rolls of the 5nm equipment and on to their brand new 3nm lines). Watching the presentations, it is quite obvious that Intel is isolated off on its own, still lying and hiding stuff, while AMD is being open and making lots of real progress with a WHOLE LOT of key partnerships. https://liliputing.com/2021/01/amd-launches-ryzen-5000-mobile-chips-for-ultralight-and-gaming-laptops.html What follows is broken out by Lilliputing, TechRadar Tom's Hardware and others --- I just got it from all over the place. Some part numbers hold the same place in the stack but have increases in core counts and base core frequency due to chiplet improvements that fire off improvements in processing power even if the part number is identical to last year's product. OR, we can base our second try at analysis at the chiplet level for greater clarity --- the base chiplets all got better, they run faster and they can do more per clock cycle. Just a few months after launching the first Ryzen 5000 desktop processors based on Zen 3 architecture, AMD is bringing Zen 3 to laptops with the launch of two new lines of Ryzen 5000 Mobile processors. As usual, the new chips are divided into 15 watt U-series processors for thin and light laptops and higher-power models designed for gaming laptops and/or mobile workstation PCs. http://https://liliputing.com/wp-content/uploads/2021/01/ryzen-5000_00-700x465.jpg Let’s look at the high performance chips first, because they’re a little more straightforward. AMD is offering these processors at 3 power levels. Ryzen 5000H processors have a 45 watt TDP, while Ryzen 5000HS versions are lower-power models with a 35 watt TDP and lower CPU frequencies and Ryzen 5000HX chips can exceed 45 watts. All of the new H-series processors are 7nm chips based on AMD’s Zen 3 architecture, with top of the line models featuring 8 cores, 16 threads, and support for frequencies as high as 4.8 GHz. MD’s new Ryzen 5000U processors are a little murkier because they’re not all based on Zen 3 architecture. In a nutshell, if you see a model number that’s a round number, you’re looking at a Zen 3 chip. If you see an odd number, it features the same Zen 2 architecture as last year’s Ryzen 4000U processors. That means the Ryzen 5 5600U and Ryzen 7 5800U will bring the biggest gen-over-gen boost in performance and energy efficiency. But it doesn’t mean that the other chips are identical to their predecessors. AMD has increased the core count and CPU frequencies, so you should see at least a modest performance bump. In the first half of 2021 AMD also plans to launch Ryzen PRO 5000 Series mobile chips for business and enterprise laptops, and all told the chip maker says more than 150 notebooks with Ryzen 5000 Mobile processors are expected to ship this year. Name Cores / Threads Base freq Boost Freq Cache TDP Ryzen 9 5980HX 8 / 16 3.3 GHz 4.8 GHz 20MB 45W+ Ryzen 9 5980HS 8 / 16 3 GHz 4.8 GHz 20MB 35W Ryzen 9 5900HX 8 / 16 3.3 GHz 3.6 GHz 20MB 45W+ Ryzen 9 5900HS 8 / 16 3 GHz 4.6 GHz 20MB 35W Ryzen 7 5800H 8 / 16 3.2 GHz 4.4 GHz 20MB 45W Ryzen 7 5800HS 8 / 16 2.8 GHz 4.4 GHz 20MB 35W Ryzen 5 5600H 6 / 12 3.3 GHz 4.2 GHz 19MB 45W Ryzen 5 5600HS 6 / 12 3 GHz 4.2 GHz 19MB 35W Ryzen 7 5800U 8 / 16 1.9 GHz 4.4 GHz 20MB 15W Zen 3 Ryzen 5 5600U 6 /12 2.3 GHz 4.2 GHz 12MB 15W Zen 3 Ryzen 7 5700U 8 / 16 1.8 GHz 4.3 GHz 19MB 15W Zen 2 Ryzen 5 5500U 6 /12 2.1 GHz 4 GHz 11MB 15W Zen 2 Ryzen 3 5300U 4 / 8 2.6 GHz 3.8 GHz 6MB 15W Zen 2 Well, this is about it for AMD 7nm as a lithography generation --- next year we are going to be doing a mix of 5nm and early 3nm mixed in with some existing old 7nm. Why such a mix? Some of 5nm won't be completely ready right off the bat and a mixed lithography bag is a likely occurrence in 2022 as we work through the global wafer shortages from the two biggest suppliers. Plus, please remember the new THEY at AMD (the new AMD Xilinx combined) will be working out the fine details of 3nm 20 lay-down layers vs 5nm 14 lay-down layers AND the blending in of the brand new programmable FPGA and AI layers in as well. Lots to do, lots to do ........ ;D |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/12/21 at 21:59:20 Why is is so hard to use AI to speed up "everything" ....... AI only works for speeding up the multiple repeats of a fixed task. AI has to be written specifically to include the AI engine, the rest of the processor data flow and the application and the task and the OS system. Intel has to keep an entire library of pre-learned solutions out on the web and your device has to go get the appropriate solution that it needs at the very best internet speeds possible in order to get anything useful done before your local task at hand is actually over. This will get better when STANDARD LEARNED MODELS for using AI in common scenarios exist, which will be a while in coming. AMD Xilinx will also be able to SELF-program their internal FPGAs to learn a repeated task directly, something that once again perhaps requires ONE quick trip home to the base library to set it up followed by some idle time for calculation. The ability to then keep the lasting customized changes locally on the PC and to use them automatically will be a lasting advantage to the AMD Xilinx system. Real AI is hard, but it is coming ....... =================================================== Inadequate laptop cooling systems are still a big big unaddressed issue ....... (CPU temps running up at 105oC and up are pretty much a standard thing now) Silicon availability and other supply chain issues are still a big unaddressed issue ....... AMD changing up their entire world twice a year creating lots of obsolescence scrap is an unaddressed issue for the desktop box and the motherboard builders. Right now the very high PC demand is easily eating up the obsoleted units at full price, but this will not always be true. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/13/21 at 15:22:12 https://arstechnica.com/gadgets/2021/01/after-corporate-blunders-and-setbacks-intel-ousts-ceo-bob-swan/ After corporate blunders and setbacks, Intel ousts CEO Bob Swan Intel is replacing its chief executive Bob Swan after a series of manufacturing setbacks and competitive blunders that lost the veteran Silicon Valley company its crown as the top US chipmaker. Swan, its former finance chief who held the top job for just over two years, will be succeeded on February 15 by former Intel veteran Pat Gelsinger, who is currently chief executive of VMware, the infrastructure software group. Gelsinger had previously spent 30 years at Intel, including time as its chief technology officer, before he left just over a decade ago to join VMware’s former parent company EMC. “Pat is a proven technology leader with a distinguished record of innovation, talent development, and a deep knowledge of Intel. He will continue a values-based cultural leadership approach with a hyper focus on operational execution,” said Omar Ishrak, independent chairman of the Intel board. “I am thrilled to rejoin and lead Intel forward at this important time for the company, our industry and our nation,” said Gelsinger. What they don't stress is that Bob Swan was only experienced as a lead beanpicker at Ebay and Bob Swan simply could not run a technical manufacturing based company he did not bother to understand. Swan also actively fired all the Intel internal potential challengers to his position. This partially explains why Intel is so very weak right now. Major Intel investors drove this most recent removal of Bob Swan, as they saw their investments totally crashing inside six months if nothing radical was done to get Swan out of Intel. Bob Swan had no real plan for Intel's future, and had no "outside partner" that was actually willing to run these more modern chipsets for Intel. :P Question now becomes "Is it even possible for Pat Gelsinger to steer Intel away from all the sharp rocks on the beach before the bow strikes ground?" Bob Swan fired dozens and dozens of critical people that he thought that Intel didn't need, missing people now critical to turning the ship around before it hits the rocks. Pat Gelsinger is having to try to rehire these people, some of which do NOT want to come back to Intel due to past Intel mistreatment. Pat may be forced to simply try his best to gently hit soft sand at low tide instead of hitting on those sharp nasty exposed rocks ....... the next available high tide will then mebbe lift him up high enough to float the boat and give him a do over. ;D Pat will immediately bump into the embedded new Intel culture of lying to stockholders and defrauding their user customers, cheating on benchmarks, BSing the computer press, etc. etc. etc. Wholesale replacement of the Intel Marketing Department may be his first task as CEO ....... as moving them out completely may well be his first significant accomplishment, one that will signal "real change" to the Federal and EU regulators who are looking hard at Intel right now. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/15/21 at 15:15:44 https://www.globaltimes.cn/page/202101/1212799.shtml#:~:text=RISC%20stands%20for%20the%20Reduced,fifth%20generation%20of%20the%20RISC.&text=Geng%20referred%20to%20the%20open,RISC%2DV%20attracting%20more%20developers. https://www.amazon.com/Iconikal-Rockchip-Computer-Processor-1866MHz/dp/B0868WSTXH http://https://www.notebookcheck.net/fileadmin/_processed_/f/7/csm_iconikal_0393_a243e1c2ff.jpg Rockchip ships a new four core A53 single board computer. $18 Comes with read out display, charger and micro USB for the operating system. This is like twice as much stuff as a Raspberry Pi comes with at half the cost (but the Pi has four much stronger A-72 cores). Beagleboard ships first RISC-V single board computer. $150 Both are a good bit less able than a Raspberry PI single board computer, with one costing less than a standard Pi costs and the other costing a good bit more than a standard Pi costs. We will watch out for the RISC-V chipset to grow more capable and pick up some decent processor speed too. Other than a "break the ice and allow programming to proceed" this $150 RISC-V single board computer is really like 20x overpriced for what it can actually do. Nothing to watch for on the Rockchip A53 as it is a subset of ARM A53 which already has several much better computer chipsets out there already. However, the cost is right at $18 so it will make its own path forward ..... and it already has good built in I/O and storage types like a laptop possesses. Double the core count and you have a midrange phone chipset with laptop I/O or a fairly strong Chromebook chipset ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/16/21 at 10:23:12 TSMC gets board approval for at least $20.8 billion dollars to fully expand their 20 stacked layer 3nm production facilities. 3nm has been board approved now for risk production, with all the rest of the new 3nm full production lines phasing in later this year. Apple will begin the task of rolling over to 20 layer 3nm direct burn EUV lithography in bulk by the second half of this year with AMD taking over the recently vacated 14 layer 5nm lines that Apple was using. This AMD changeover will go very quickly as wafers full of the same AMD chiplets are relatively simple to run compared to a full complexity Apple M1 SOC wafer that the process was just doing. Pre-production development testing begins in earnest on the 1.5nm existing early production process lines. Work begins with the first 3nm customers on the new process node intended customer chip designs (Apple and Huawei). Early development and testing always involves customer designs at TSMC as nothing is real at TSMC except what the customer needs and requires. So, this year 5nm goes to 100% full scale production levels with all the 5nm buildings being stuffed with work --- Apple rolls down to 3nm and AMD rolls into filling the 5nm lines. Many of the most modern 7nm lines will now be freed up for TSMC's 6nm conversion which will then be filled with phone customers and possibly some Intel work if Intel can finally get over itself and commit the funds to TSMC. Inside 2 years, 1.5nm chipsets and chiplets will be rolling out of newly built TSMC buildings ...... what will Intel be doing ....... GOOD QUESTION, other than jest sitting there on their thumbs being 4 lithography generations behind ???? An additional 10 billion just got added to this TSMC pile as demand from GOVERNMENTS is having to be factored into this build-it money. Yes, a secured military production facility is going into ARIZONA, USA by US fiat. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/17/21 at 18:54:56 Rumor has it that Intel has had to give up completely on Optane memory --- that when the Intel warehouses go empty there will be no Intel Optane anything ever again. Or, let's be more precise --- Intel has no partner willing to make the stuff for them any more. Their ex-partner may have removed all Intel rights of use to Optane tech completely as Intel never paid their old bills over use and scrap and other costs. Acting in bad faith can really cost you greatly, as Intel will soon learn. ::) I wonder how many other items that Intel "invented" in the last few years were equally dubious and potentially just more sorry Intel BS ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/17/21 at 19:17:59 https://arstechnica.com/gadgets/2009/03/amd-intel-engaged-in-x86-licensing-staredown/?itm_source=parsely-api Bob Swan also did something else that was very very stupid, he has started a new proceeding against AMD over the licensing of the original x-32 bit X86 instruction set, and he did this stupidity purely as a FUD action to toss some shade on AMD right before CES. AMD likes this very stupid FUD only action very very much, however, because AMD owns more key AMD x-64 bit patents, multi-CPU patents and "fusion" patents (joining CPU and GPU on the same silicon), way way way more current stuff than Intel does, stuff that Intel has been using without a discrete license. AND PLEASE REMEMBER THAT AMD HAS ALREADY WON 2 PREVIOUS COURT CASES THAT GIVE AMD THE ONGOING RIGHTS TO USE THE BASIC "Intel 32 BIT x86" TECHNOLOGY THAT BOB SWAN WAS DISPUTING. VIA also has ongoing rights from those same cases to use the original x86 tech and now so do the Chinese that own the ghost of VIA right now. But old Ebay bean picker Bob didn't know all that ...... he hadn't been around that long. Patent suit history between AMD and Intel AMD does not believe AMD has breached the cross-licensing agreement, and the company considers Intel's allegations of a breach to be in bad faith. If Intel's filing does turn out to be in bad faith, then AMD claims that this gives them "the right to terminate Intel's shared rights and licenses from the AMD side, while retaining the Company's [AMD's] rights and licenses from the Intel side under the Cross License Agreement" as is stated in court judgments made twice previously. Once AMD rolls out the conjoined Xilinx chipset they won't be using any oldstyle 32 bit x86 technology at all anyway, so now they have an additional string to their bow. New AMD patents have been filed already on the AMD-Xilinx conjoined tech stuff and these patents have been accepted by the US Patent Office. Xilinx has never allowed Intel any form of use license for AI or FPGA, ever. Intel is potentially going to lose out very very very big time due to this very very stupid Bob Swan action if Intel pursues it. Did Swan even realize that he is trying to enforce an obsolete 32 bit chip system patent that isn't worth anything today and that AMD actually invented and owns the patent rights and instruction set copyrights for all the modern 64 bit instruction set and all the multi-processor system patents and the patent rights for building multiple CPU chipsets and the patents for using chiplets to make up a processor ????? Yes, AMD owns the patents for the use of chiplets, and just calling them tiles does not give Intel the right to use the tech AT ALL. Xilinx invented and owns every sneaky AI booster tech that Intel uses to juice up their out of date processors, so Intel will lose out on that stuff big time, too. Bob Swan was REALLY STUPID taking this action and Bob Swan really did not know enough to even try to run a tech company like Intel. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/18/21 at 12:44:07 https://siliconangle.com/2021/01/16/new-ceo-pat-gelsinger-must-channel-andy-grove-recreate-intel/ http://https://d15shllkswkct0.cloudfront.net/wp-content/blogs.dir/1/files/2021/01/Pat-Gelsinger-Dell-Tech-cube-intel-background.jpg New CEO Pat Gelsinger must channel Andy Grove and recreate Intel We believe that Intel has no choice. It must create a deep partnership with a semiconductor manufacturer aspiring to manufacture on U.S. soil, and focus Intel’s resources on design. In this Breaking Analysis, we’ll put forth our prognosis for Intel’s future and lay out what we think the company needs to do to, not only maintain relevance, but regain the position it once held as perhaps the most revered company in tech. Peak PC volumes signaled the top of Intel's reign http://https://d2axcg2cspgbkk.cloudfront.net/wp-content/uploads/Breaking-Analysis_-Pat-Gelsinger-Must-Channel-Andy-Grove-and-Recreate-Intel-1.jpg But in the all-important volume semiconductor manufacturing game, PCs are no longer the king. Collectively, Arm chip units surpassed 20 billion worldwide last year. Comparing Arm chips shipped with x86 PCs is not apples-to-apples, but the numbers speak for themselves: The wafer volume for Arm chips dwarfs that of x86 by a factor of 10 times. Back to Wright’s Law. How long will it take Intel to double their wafer volumes? Gelsinger understands this dynamic probably better than anyone in the world – certainly better than we do. The answer is NEVER ....... as ARM grows in volume x86 declines by the same ratio. Moreover, if you look at the performance and price-performance of Arm compared with x86, the picture continues to favor Arm, quite dramatically. Arm design to production cycles are faster, the technology is climbing the learning curve faster and riding the cost curves down at a more accelerated rate. Arm encroaches on PCs by growing 10x faster As you look out to the future, the story for Intel’s PC dominance also does not look promising. http://https://d2axcg2cspgbkk.cloudfront.net/wp-content/uploads/Breaking-Analysis_-Pat-Gelsinger-Must-Channel-Andy-Grove-and-Recreate-Intel-5.jpg http://https://d2axcg2cspgbkk.cloudfront.net/wp-content/uploads/Breaking-Analysis_-Pat-Gelsinger-Must-Channel-Andy-Grove-and-Recreate-Intel-10.jpg No, Gelsinger is a visionary with a deep understanding of tech, architectures, trends, markets, people and society. He’s a dot connector. And he loves Intel– he’s a legend at the company where he spent 30 years. A major strategic partnership with TSMC or Samsung is needed Here’s what we strongly believe. We think Intel must do a deal with TSMC or Samsung – perhaps a joint venture or some type of innovative structure that both protects its IP and secures its future. Both of these manufacturers would love to have a stronger presence in U.S. markets where Intel has many manufacturing facilities. They may even be willing to take a loss to partner more deeply with Intel. In our view, Intel needs to get to 5nm within two years to remain cost-competitive and defend its base. It can’t do this in our opinion without a strategic manufacturing partnership. If it tries to go it alone, the costs of introducing next-generation technologies will be so expensive it will bankrupt the company. How’s that for paranoia? To be clear, we believe that core Intel should get out of the business of manufacturing semiconductors. Set up a JV and spin the venture off so you can profit but focus the core company on design. That will allow Intel to compete better on a cost basis with AMD, defend its data center revenue — and fight the good fight in PCs, better preparing for the coming onslaught from Arm. Then go on offense and remake Intel from the inside Intel should put a laser focus on reducing its cycle times and unleashing its designers to create new solutions. Let a manufacturing partner who has the learning-curve advantages enable Intel designers to innovate and extend ecosystems into new markets. Autonomous vehicles, factory floor use cases, military, security, distributed cloud, the coming telco explosion with 5G, AI inferencing at the edge. Just think about the opportunities for Intel in 5G, which the company has pegged as an $18 trillion opportunity. Bite the bullet and give up on yesterday’s playbook and reinvent Intel for the next 50 years. That’s what we’d like to see and that’s what we think Gelsinger will conclude when he channels his mentor. We sincerely wish the best for Pat, the people at Intel and the continued success of a great American company. =================================================== AMD already has a strategic production alliance with TSMC that is holding firm. TSMC gets first crack at any AMD wafer requirements, with Samsung being the alternate supplier. Apple keeps the same mix of primary/secondary suppliers for the exact same reasons. This week Samsung announced an IP sharing plan with AMD over processor construction, graphics and other base technologies that Samsung needs to compete in PC class equipment. AMD gets more Samsung 5nm production wafer space if TSMC cannot supply all that is needed, a key consideration that AMD needs over the next few years. AMD got there first with making the alliances and partnerships, and Intel is still sitting off all by themselves ....... Intel is consistently a day late and a dollar short in their partnership efforts. Remember, Samsung and TSMC are both 5x bigger than Intel is now and they are growing while Intel is crumpling as we all watch. Intel must first and foremost start to honor all old contracts and relationships and actively change their image away from the company that sues and screws over every last one of its old business partners. =================================================== Pat Gelsinger is simply aghast at what Bob Swan has done, letting go of so many key personnel as "unnecessary" --- Pat has already begun efforts to try to hire back some of these "forced to retire" folks. Some will come, some say they are done with Intel's poor treatment of loyal employees and will not return. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/20/21 at 20:26:42 http://https://www.notebookcheck.net/fileadmin/_processed_/5/4/csm_2021_01_04_16_53_13_PassMark_CPU_Benchmarks_AMD_vs_Intel_Market_Share_____Mozilla_Firefox_77642db818.png Updated graphs say it all. AMD could sell more if it could make more, and that simply requires relatively MORE CHIPLET WAFERS from TSMC and SAMSUNG so it can expand production. 7nm production is "all full up" at this stage of things as everyone is busy rolling over to 5nm very soon and all of 7nm must be all run out ASAP before the 7nm equipment is taken out of service to be refurbished into the new TSMC 6nm production lines. Dancing on down the nm's is the name of the game now ......... AMD is asking TSMC for more 3nm, 5nm and 6nm chiplet production wafer allocations ASAP. Samsung is also being asked for more production wafer allocation on the 5nm AMD style chiplet wafers. AMD is in a prime spot to make some hay, but it needs a lot more chiplet wafers to do that. ALL of the new lithographies are better than what AMD is using now, so AMD should be able to get something they can readily use in volumes big enough to take another chunk out of Intel's arse. AMD NEEDS MORE CHIPLETS !!!!!! AMD needs to put in greatly increased requests for initial 3nm production right now if they wish to get any significant volume of 2021-22's 3nm chiplets for a "top of the line" premium product or two...... All of TSMC's future CAPEX is going 100% into the brand new (current tech) 3nm and 2nm production facilities with all these new scanner orders already booked at ASML in Holland. All the new 5nm scanners that TSMC ordered have been built already and are being delivered right now ....... (Apple takes first position, then rolls on down to 3nm during the upcoming year) So, Intel and AMD are now at war over production capacity at TSMC and at Samsung with the winner taking it all as the loser can't build enough of anything to even try to take back any of their lost market share ...... Several smaller phone suppliers will be left out in the cold during this struggle as they will have to find some chip supplier somewhere to make their stuff for them. =================================================== This is a visual on AMD's growth in graphics which is very nice but not even close to the same order of magnitude to AMD's CPU wins last year. http://https://preview.redd.it/07scsqt0yqc61.jpg?width=960&crop=smart&auto=webp&s=edf7d20f8da075b5c09855d05f93b92e0a428fe5 |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/22/21 at 12:40:57 https://liliputing.com/2021/01/intel-ends-2020-on-a-strong-note-thanks-to-record-notebook-sales-but-7nm-chips-arent-coming-until-2023.html After years of stagnant or declining growth, global PC shipments were up in 2020 due largely to an increase in notebook sales as people hunkered down to work and school from home during the global pandemic. And it looks like Intel rode that success to better-than-expected revenue. In the company’s latest earning report, Intel says “PC unit volumes” in the fourth quarter of the year were up 33 percent over the same period in 2019, “led by record notebook sales.” But that apparently wasn’t good enough for investors, because the company’s stock price plunged again this morning. One possible reason? The company faces some serious challenges in the coming years. http://https://liliputing.com/wp-content/uploads/2021/01/intel-silicon.jpg Specifically, while Intel has finally worked out the kinks in its 10nm manufacturing process and has migrated most of its chips to the new node, the company isn’t expected to make the move to 7nm until 2023, which is 6 years later than Intel had originally hoped to hit that target. Rival AMD, meanwhile, has already transitioned to 7nm for its latest laptop and desktop processors and is moving down to 5nm this year. While you can’t directly compare one company’s 7nm technology to the other’s, the point is that Intel has been much slower to move from one node to the next in recent years. Intel is staying 2-4 lithography nodes behind the leaders at this point in time. In fact, in order to meet its 2023 deadline, Intel’s incoming CEO Pat Gelsinger told investors that “it’s likely that we will expand our use of external foundries for certain technologies and products,” but that the company does still expect that the majority of its chip manufacturing will likely be done in-house rather than by outsourcing jobs to third-party foundries like TSMC. Investors call BS on Intel for not being able to move to 7nm before 2 more full years have elapsed. This is 2+ years of getting their faces rubbed in faster AMD chiplets that are 2-3 lithography generations improved over the best Intel can offer. And yes, Intel is liable for stockholder fraud for NOT SAYING THIS KNOWN FACT EARLIER and for hard selling their stock based on things Intel management knew were never going to happen in the time frame that was specified in the stock holder communications. It must be hard to be Pat Gelsinger, starting out by having to apologize to his own stockholders for what his predecessors did to defraud them. =================================================== Intel is currently noted for its somewhat secretive use of AI to speed up its aging processors (make them appear to work faster on certain tests). This is not news, the phone boys have been doing this for most of a decade now. Huawei was noted for doing a software update on their phones and then showing test results on last year's phones that exceeded what the competition could do with their newest chips --- Huawei got blasted for cheating at the time, but it was just the AI tricks hitting the benchmarks for the very first time. I find it funny that all phone chips (even low end ones) now have a 0.8 Trillion Operations per second AI accelerator tucked into all of them, even a cheap MediaTek lowrange phone chip. The better Qualcomm chipsets are all running OVER 12 Trillion OPS on their AI packages ...... the high end best from Qualcomm is currently running 24 Trillion OPS. Apple is only at 4 Trillion OPS on their AI as Apple really isn't into appearances type cheating like the rest of the phone boys. Some open cheat proof benchmarking of this sort of action is needed ---- and some unification of the extensive on-line libraries that back up this new usage form. If the trick works, then use it openly for more than just a benchmark test. Don't lie about it ....... use it openly. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/23/21 at 00:21:23 Confirmed Rumors (from Texas, supposedly) TSMC is building a 5-3-2-1nm plant in Texas. Arkansas is mentioned as another TSMC site in some rumors. Samsung is building a 5-3-2nm plant in Texas. Both plants are to be complete inside 1.5 years with rolling updates to equipment planned going forward from there. Both companies are bringing along their own favorite suppliers to support this move. Think of it as a glorified "Assembled in the USA" situation which is close enough to true. Plant CapEx is around 11 billion dollars for each plant initially as the ASML lines in them are very expensive. Enough land was purchased for the growth and expansion of buildings. Texas is offering the best incentive packages so Texas gets the plants we know about now. Arkansas was also mentioned as a potential location. These plants are "any customer" plants (not dedicated). =================================================== BTW, Intel has started saying this week that they now plan to make their own chipsets on their own Intel designed and built custom equipment in their own existing Intel facilities. So far Intel gets sticker shock whenever they go to buy a processor built off somebody else's equipment. Pat Gelsinger finds it cheaper to use his own existing manufacturing line hardware and to hire back some of his old experts who knew how to make up the lines and how to run them (and how to tweek the processors to work on them). Pat is also finding that Bob Swan had already burned all of Intel's "potential partner' bridges, so Pat has no choice but to go do it all on his own. ================================================== ......... speaking of getting sticker shock on processors ........ Why build the TSMC and Samsung low lithography plants right now? Avoiding the newest tariffs and taxes being put in place by our new political controllers will pay enough to make it worthwhile for all the parties concerned. To make this game work, the incept date of the plants had to clearly precede the tariff legislation it was designed to avoid. So far this seems to be in alignment. Democrat run congress is busy chasing Trump at the moment ..... but they could swing back into tax & spend mode at any time. =================================================== There is a new graph line for ARM computer processors that is needed on all these graphs. Using Apple M1 for the sole contributor at this time, the ARM line takes 15-20% market share when folks finally pop the new ARM market share line into the rest of the future market share graphs. Apple M1 represents the first section of this green wave that hits starting this year. Apple is going to release the 12 core version of their chipset this year, covering all laptops and the lower side of Apple desktop PCs, with a 20 core version planned for later on. Qualcomm and Huawei will contribute another smaller chunk this year and next, after that all the phone boys will be in there swinging their hockey sticks at Intel, AMD and at each other's heads ...... By then the market share hits landing on x86 will be larger and more noticeable. http://https://d2axcg2cspgbkk.cloudfront.net/wp-content/uploads/Breaking-Analysis_-Pat-Gelsinger-Must-Channel-Andy-Grove-and-Recreate-Intel-5.jpg |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/25/21 at 19:48:51 TSMC and Samsung are getting the riot act read to them by the US and the EU governments. Build local plants in our countries that we can defend. Chipsets are a national defense PRIORITY, you can charge us a lot of money but you MUST build us enough chipset manufacturing capacity for us and you must do it in defensible locations inside our boarders. The bullshite about running out of chipsets STOPS, ASAP. If you need money, tell us how much ....... Folks, Intel will never heal itself fast enough in this environment. AMD is not likely going to get enough chiplet wafers short term to keep up with their spiking demand either ....... TSMC and Samsung have the will and the knowledge and the supplier base to drop plants wherever they are required. Nobody is going to fund Intel for a government sponsored anything after that last supercomputer fiasco so get over it Pat Gelsinger, you old alma mater is water over the dam as of these latest announcements. Nobody wants "Intel Inside" any longer. China has a military conquest plan for South Korea and Taiwan that in one fell swoop would give them all of the current processor manufacturing in the entire world. Look to see Biden's Administration continue the hard line with China that was set up under the Trump Administration. Also remember, ASML the Dutch company that makes the 3nm 20 layer EUV processor lines can get told by the EU to stop sending scanners to people who don't play well with others. What happened to Huawei can happen to you, too. ================================================== https://www.reuters.com/article/us-huawei-smartphone-sale-exclusive/exclusive-chinas-huawei-in-talks-to-sell-premium-smartphone-brands-p-and-mate-sources-idUSKBN29U0OU Rumor is from Reuters (who hates rumors) China denies all of it, so it must remain a rumor for the time being. But the Huawei Honor brand just got sold off, so it is clear that something is likely happening as we speak. (Reuters) - China’s Huawei Technologies Co Ltd is in early-stage talks to sell its premium smartphone brands P and Mate, two people with direct knowledge of the matter said, a move that could see the company eventually exit from the high-end smartphone-making business. The talks between the world’s largest telecommunications equipment maker and a consortium led by Shanghai government-backed investment firms have been going on for months, the people said, declining to be identified as the discussions were confidential. Huawei started to internally explore the possibility of selling the brands as early as last September, according to one of the sources. The two sources were not privy to the valuation placed on the brands by Huawei. Shipments of Mate and P Series phones were worth $39.7 billion between Q3 2019 and Q3 2020, according to consultancy IDC. However, Huawei has yet to make a final decision on the sale and the talks might not conclude successfully, according to the two sources, as the company is still trying to manufacture at home its in-house designed high-end Kirin chips which power its smartphones. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/26/21 at 08:19:10 https://liliputing.com/2021/01/the-first-ryzen-5000h-laptops-are-now-shipping-and-their-really-fast.html https://www.tomshardware.com/news/amd-ryzen-5000-mobile-cezanne-soc-architecture-deep-dive-zen-3-powers-into-notebooks https://www.anandtech.com/show/16446/amd-ryzen-9-5980hs-cezanne-review-ryzen-5000-mobile-tested http://https://liliputing.com/wp-content/uploads/2021/01/5000h_02-700x392.jpg The first Ryzen 5000H laptops are now shipping (and they’re really fast and totally for real) by Brad Linder Posted on January 26, 2021 at 10:52 am The first Ryzen 5000H laptops are now shipping (and they’re really fast) AMD unveiled the Ryzen 5000 Mobile processor lineup earlier this month, and now the company says the first laptops with 35+ watt Ryzen 5000 H series chips hit the streets starting today. Ahead of launch, the chip maker sent Asus ROG Flow X13 thin and light gaming laptops with Ryzen 9 5980HS processors to a number of tech reviewers, and early reviews indicate that AMD’s flagship “Cezanne” processor largely delivers on AMD’s promises: users can expect double-digit performance increases in single-core and multi-core workloads. That’s due to a move from Zen 2 architecture to new Zen 3 CPU cores, a new unified L3 cache structure (not to mention double the overall cache, now at 16MB), an upgraded memory controller, and a new memory controller, among other things. As they typically do AMD slightly exceeds their advertised expectations ..... I hate it that the newest chipsets have such a pricing premium attached to them during distribution, but supply and demand sets the final prices in during any sort of scarcity situation. Sux, but it is what it is ...... we got severe scarcity showing up all over the place. Tom's Hardware has now published their deep dive on the Cezanne series too (yes, AMD has all of their lineup slots refilled with processors built with the new Cezanne generation chiplets). So far Tom's finds no downsides in the new chiplets and finds performance is slightly exceeding the advertised amounts and Cezanne is clearly better than Intel's latest AI pumped up offerings. Anandtech has done their deeper dive as well. There is a comprehensive list of code names with the resulting sources and lithographies for those who are interested. Since there is higher wattage now on tap (with higher core temperatures to go along with it) there is an analysis of laptop cooling systems and active temp control set points. http://https://images.anandtech.com/doci/16446/Test1_678x452.jpg yes, that is melted solder for the "thermal paste" -- that is how high temp chipsets roll now-a-days Now, let me digress a bit. AMD can roll a new improved architecture in one generation and then roll chiplet improvements several times during a generation. Then there are the much bigger lithography generational roll outs, which are spaced out several years apart. Reminds me of a Civil War Era LeMatt revolver, if the 36 caliber bullets didn't get you the 42s did, and if that failed the center pin 20 gauge shotgun barrel was brought into play to lay the big BaDaBoom down on your arse. :o Later this year we get to do it all over again, this time with 14 layer 5nm chiplets ........ :o How much real improvement has AMD done in the last 3 years? 2,100% real improvement. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/27/21 at 19:25:52 Intel is doing something curious right now, they will not sell you an Intel video card as a separate sale. It must come as bundled as part of a brand new complete computer by one of their tame suppliers. Why? Because it will not work ......... On an AMD based computer system On an ARM based computer system On a Linux System On an Android System On any system not set up complete with all the secret Intel drivers and libraries required by the card to get the secret hidden AI to run fast. On your just bought brand new box stock Intel system if you should change the OS (or the system image) in any appreciable fashion. Intel and their new processors and graphics cards have some very very serious issues they need to work through. Microsoft has some serious issues with keeping up with your Windows 10 license vs "Intel approved hardware". The default answer in this situation is "you just broke it and need to go buy a brand new Windows 10 license". Surprise, the new Windows instance won't work after you try to install your new operating system ...... you are not an approved installer and you did not provide the secret sauce. Pat Gelsinger has severe amounts of PR work to do. No PC enthusiast will put up with this level of BS. Apple will get some serious amounts of brand preference out of this situation as will AMD based stock units. Apple and AMD just work ..... using simple "normal for the OS" drivers, no secret sauce or AI driver libraries required. =================================================== Now that the 3 new low lithography plants have been announced, Intel is saying by 2022 they will be in production at 3nm on "some products" using these plants. The news of the 3 USA low lithography plants is good, but by the time they are built they will be over booked very very badly. Nobody wants any old 14nm Intel stuff any longer, Intel has stretched 14nm so far now that you can see light through the holes in the product. Nobody trusts Intel 10nm, because it requires a lot of secret AI sauce slathered on it just to be somewhat equivalent performance-wise. Nobody thinks Intel 7nm is ever going to work. AMD at 5nm will be OK, but AMD at 3nm is likely to be completely ass kicking in ways we don't even understand yet. 2022 isn't that far away ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/27/21 at 23:07:31 SCARY RUMORS AGAIN We got us the first moves of a international computer systems war going on. TSMC and Samsung are getting reeled in hard. Huawei, Oppo and SMC and all the other Chinese clone companies are getting cut off at the knees. Red Hat has just shut down all support on their open source clone Centos. Google has just shut off "free and full" functionality updates to Chromium Open Source Project. Google has shut off sharing new features to the Open Source Android flavors (as created by various distros and vendors) and will only support their own official version of Android going forward. American computing companies are getting told to only support American and ally uses of official products. They are getting told to functionally "update break" and then intentionally shut down all the bogus "rebranded" software systems, most specifically any that include modern proprietary features that were sneaked in on the sly by the orientals running the clone. Prosecutions of various known Chinese and Russian hackers are rolling into the courts. Three brand new fairly large processor plants are being built on US soil at all possible speed. Two by TSMC and one by Samsung. All of them are 5-3-2nm plants. THE ENTITY LIST IS GETTING LONGER BY THE WEEK. THIS IS NOT TRUMP --- THIS IS US MILITARY, FBI AND CIA IN ACTION AGAINST THREATS AGAINST THE AMERICAN REPUBLIC. COVID 19 IS NOW BEING SEEN AS AN "ORCHESTRATED ATTACK" AS THE NEW STRAINS POP UP INTERNAL TO THE USA WITH NO TRANSMISSION "PEOPLE" VECTOR BEING IDENTIFIED. THE OLD RUSSIAN METHODS OF NEUROTOXIN ATTACK IS BEING SEEN AS A REPEATED POTENTIAL PATTERN FOR THE ORIGINS OF THESE NEW COVID STRAINS HERE IN THE USA. :o wow, I am scaring myself |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/28/21 at 10:50:28 With the automotive plants shutting down due to no chip supply --- chip allocations will begin immediately. Auto plants get priority one. PC and laptop chipsets get priority two. Phone chipsets get third priority. (keep your old phone, put a battery in it) Construction of domestic USA chip production plants by TSMC and Samsung is a crash priority 1 issue right now as NOTHING RUNS IF THE CHIPSET SUPPLY ISN'T THERE. Right now the world has run out of chip making capacity. Apple has a quarter of the 5nm capacity at TSMC locked up in their 5nm roll out, as soon as that is released AMD locks it up for the next two years. Phone companies have divvied up all the rest of the 5nm capacity. 3nm roll out is immediate and critical at TSMC and at Samsung. This relief is needed NOW. In line with this, TSMC reports Intel has signed contracts to build out their line at 3nm on TSMC processor lines. This means they must design to TSMC criteria and must do all test runs at TSMC on the same lines where the stuff will be built. 2022 is the build year in question, so early prototyping is a NOW sort of thing. Look to see Intel start leaking some better Core i5, i7 and i9 results starting NOW. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by jcstokes on 01/28/21 at 15:06:23 Whats all this scary stuff mean for some poor sap running Linux Sonya on a 12 or13 year old HP desktop? Aw, and Win 7 somewhere still on it. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/28/21 at 15:41:25 It means you are fine and will continue to be so. Linus and the boys are pretty much unaffected as they do not deal in hardware as a primary business. Whenever Intel gets organized and publishes a standard, Linux will likely incorporate the new Intel stuff into Linux as they always have done. Intel's issue is SECRECY as Linux does not incorporate secret anything --- they call that stuff proprietary hardware and proprietary software blobs and simply won't touch it. ==================================================== The scary stuff is military people carry guns and historically tend to shoot at each other over stuff like this when it all gets tense. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/28/21 at 16:03:43 https://liliputing.com/2021/01/idc-huawei-smartphone-shipments-tank-but-the-market-is-up-overall.html Huawei smartphone shipments tank, but the market is up overall First up, there’s been a shakeup in market share. Huawei is said to have shipped 32.3 million phones in Q4, 2020, which sounds like a lot until you realize that figure is down from 56.2 million phones in Q4, 2019. That’s a 42.4 percent decline, and it means Huawei is now the world’s fifth largest phone maker rather than the third. The US started imposing trade restrictions on Huawei during the Trump administration, but it’s unclear if things will change under the Biden administration. So with Huawei cut off from some of the technology it needs to continue making smartphones, rumor has it that the company is considering selling off its flagship Mate and P smartphone brands. Volumes Market Share Volumes Market Share Year-Over-Year Change Apple 23.4% 73.8 19.9% 22.2% Samsung 19.1% 69.5 18.8% 6.2% Xiaomi 11.2% 32.8 8.9% 32.0% OPPO 8.8% 30.6 8.3% 10.7% Huawei 8.4% 56.2 15.2% -42.4% Others 112.4 29.1% 107.1 28.9% 5.0% Total 385.9 100.0% 369.9 100.0% 4.3% |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/28/21 at 21:15:56 Chinese container ships unload Chinese manufactured goods, REFUSE to load the return shipments of agricultural supplies, say they will only load empty containers for the return trip to China. This has been going on for a month now. This is an escalation of the trade conflict. America is now under an unannounced vegetables and grain EMBARGO from China. No reason has been given out so far. Supplies of cargo containers are drying up in America as the one way exit of the world's supply of cargo containers bleeds American commerce dry of containers. If you want a container, you have to pay to have it shipped over the pond to you. You pay the doubled cost of shipping. Not shooting bullets at each other yet, but things are worsening. Huawei smartphone shipments tank as Huawei sells off phone product lines to near eastern partners that are not currently under sanction, but the global marketplace is getting closer to open economic warfare. The ones buying the Huawei stuff may well find themselves on an entity list just like Huawei was ...... so take care who you are owned by and who you deal with. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/29/21 at 16:54:33 https://liliputing.com/2021/01/canalys-tablet-and-chromebook-shipments-hit-all-time-highs-in-q44-2020.html As PC and laptop chipsets become scarce allocated scarce resources, is anything picking up the slack? YES, Chromebooks and Tablets are moving over to fill the bill ......... http://https://liliputing.com/wp-content/uploads/2021/01/canalys-record-tablet-and-chromebook-700x394.jpg Question becomes, if they do a good job will they ever give that market share back ???? |
|
Title: Re: AMD & others --- post AMD dominance in 2020 Post by jcstokes on 01/29/21 at 17:51:24 Very interesting your last bit about Huawai, I like several hundred thousand New Zealanders have a Huawai computer modem supplied by Spark, formerly Telecom, formerly New Zealand Post Office. They are probably still the largest telco and isp in this country followed by Vodafone. This modem doesn't update by asking you if you want the updates. Huawai slithers about inside your computer attaching itself to websites, including this one. It then attempts to block you using the site until you've installed their update system. There are ways of avoiding this, but I don't like the sneaky way in which they try to do it. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/30/21 at 02:49:40 Can't speak for their business practices, but Huawei and the Chinese Military are said to speak from the same perspective, which is why they got put on the USA Entity listing in the first place. And wading around inside your network and inside your PC is a part of it, too. I can say that in phones Huawei historically cheats on benchmarks ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/01/21 at 07:29:49 Apple has a hard lock on TSMC 5nm production and hasn't turned it loose yet to move to TSMC 3nm. Apple is on a roll and can sell everything they can make, so Apple's dedicated 5nm lines keep right on rolling at full speed. AMD is on a roll too and AMD needs some more production of chiplets. AMD has made up a deeper relationship with Samsung, sharing their AMD Radeon video subsystem expertise and as payment is getting much larger 5nm and 3nm chiplet production allocations from Samsung in exchange. AMD has always had Samsung listed as their alternate production for some time now, but TSMC has never before been unable to fill orders for AMD in a timely fashion. TSMC is taking some Intel orders now on older lithography products and quite frankly TSMC is cherry picking where their production goes, seeing they have entire governments telling them do this and do that now. TSMC simply cannot make it all any longer. So, expect to see some Samsung gate all around 3nm to show up in some AMD chiplets unexpectedly soon. It is thought that TSMC needs to expand its 5nm and 3nm production capability by roughly 3x just to handle what is on their plate right now. Samsung is also growing by leaps and bounds right now as Samsung is doing the lion's share of memory production right now as well as part of TSMC's memory share as TSMC is simply too busy making chipsets for people. Intel is now functionally dropping the ball on about 25% of the world's chipset needs, stuff that they had contracted to produce at 7nm and cannot fulfill now. TSMC and Samsung need to build a bunch more capacity if Intel is dropping out of 7nm chip manufacturing as they seem to be doing. Both Intel and Global Foundry can offer you some 20nm, 16nm and 14nm production, but lead time for starting new orders is 6 months roughly. So, falling back to your old designs that have to be built at the old lithography nodes is being attempted more and more due to simple necessity, but this backup plan is shortfalling as well. Old lines 7nm historically got pulled and refurbished into newer 6nm lithography, just as one current example. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/02/21 at 03:51:23 https://www.tomshardware.com/news/amd-gpu-chiplet-patent http://https://cdn.mos.cms.futurecdn.net/T9FhGXUj2ry5v63mDwf4wf-970-80.jpg.we Breaking Through the GPU Performance Barrier using chiplets On the face of it, this move certainly makes sense. As you increase die sizes, yields tend to drop drastically due to defects in the silicon, leading to costly losses that impact prices. Semiconductor manufacturing equipment also has a reticle size limit, which essentially creates a barrier beyond which it simply isn't possible to make a bigger GPU. Coupled with the state of the process node (TSMC is currently at 7nm), there is a performance ceiling that foundries simply cannot manufacture beyond. AMD Chiplet GPU Patent http://https://cdn.mos.cms.futurecdn.net/WsExyydC7PFCAkiwBqSJtG-970-80.jpg.we However, breaking the big silicon die into multiple little ones addresses all these pain points. And as such, AMD proposes that a chiplet-based GPU just needs a simple electrical path to be tied together to look like one big GPU, rather than a bunch of little ones addressed through some form of controller. AMD Chiplet GPU Patent http://https://cdn.mos.cms.futurecdn.net/wsyBhfyajWQccC8YMJ2usF-970-80.jpg.we The beauty of this GPU model is that it remains compatible with existing coding languages, and the GPU model doesn't change from the developers' point of view. A few things will need to be adjusted in the driver, naturally, to account for the new architecture, but existing software should run without any major changes. What I like about this is the new on chip communications fabric improvements is that it may well be able to flow back into the CPU side, making it possible for resources to be shared between CPU, GPU and the upcoming FPGA/AI functions. And, having gaming level graphics boosted by programmable-on-the -fly AI all on the same 14-20 layer chipset --- hey, what is not to like? ================================================== Intel reacts to this serious announcement of a real AMD development project with BS about a 20 core Big Little with no graphics on board (Intel will use a separate GPU chipset if they ever get around to building this theoretical response). AMD posts die shots of real engineering sample parts ---- Intel responds with some nebulous BS verbiage about an not yet designed Intel big little that would be built on a lithography node that Intel admits it cannot get to run right. Which one do you think will be built inside a year from now ?????? Intel's stock price is getting simply hammered over shite like this ----- Pat Gelsinger, make the bad PR boys stop their endless lip dribbling as it is jest killing your image. ================================================== Year end results paint an interesting picture. AMD APUs were what was desired, but Intel Tiger Lake was what was available and that was what actually got bought at the end of the reporting period. Supplies of full PC chipsets from everybody went dry at the end ---- so a lot of Chromebooks got bought in the last few weeks. TSMC can no longer supply all of AMD's needs, so Samsung is getting rung into the mix on an ongoing fashion. TSMC can always get this volume back by contract terms, but only by supplying all of AMD's orders religiously on time and at the contracted price. =================================================== TSMC has already published their 5nm expansion plan (nothing else is new, really, TSMC 5nm is already all built out and it is all set up and rolling product right now --- and it is all firmly allocated to Apple and AMD and Qualcomm, and the phone boys, the guys who paid to build it). TSMC's 3nm expansion is about half built, but is already totally scheduled tight and full through 2024. Same major players paid to build this 3mn stuff and have first dibs on what will be a scarce allocated resource from the very 3nm beginning. It is already known that Intel did not buy into either 5nm nor 3nm by the original cut off time and as such INTEL HAS NO ALLOCATION LAID IN FOR ANY CURRENT EXISTING 3nm TSMC PRODUCTION LINES THROUGH YEAR 2024. Intel claims that they do have allocations though, so we shall see. We do have the two "anybody's production" 3nm plants going into Austin Texas, and perhaps that is what Intel is counting upon. And, Intel may be referring to Samsung instead of TSMC. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/04/21 at 12:51:29 Sifting Rumors (to separate the PR bull from the pure stinky shite) Several sources in Taiwan say that TSMC has cut no deal with Intel. Intel claims otherwise. Samsung is indeed currently building a small amount of low lithography graphics materials for Intel. Intel is trying to get substantial production allocations from both, but Intel has nothing to offer in exchange. All of the current 64bit x86 patents belong to AMD, not to Intel. AMD did cut a technology sharing deal with Samsung so the AMD Xylinx TSMC Samsung IBM technology consortium is still somewhat of a thing. AMD is seeking significant production allocations from both TSMC and Samsung, with the clear understanding that when TSMC has production capacity all the existing current stuff rolls back to TSMC according to the existing contracts. Things that are starting up now in the "vague zone" are a different situation though. All the new new stuff needs a design/build home, and both sources can apply for the business. Once you get the business, you will tend to keep it as the new products will be HAVE TO BE DEVELOPED IN COOPERATION WITH AND TO EXACTLY SUIT THE MANUFACTURER'S PROCESS. =================================================== NEWS FLASH as of Friday 2/5/21 TSMC acts to defend their AMD agreement, saying that worldwide substrate shortages caused them to not make the AMD orders --- not any lack of capacity on TSMC's part. TSMC says they have now fixed the substrate issues and will fulfill all AMD requirements promptly and at the contracted price, so no Samsung production is needed. CNBC reports that the substrate issues have been fixed and the Biden Administration thanks Taiwan for help in resolving Automotive chip shortages (and all the rest of the shortfalls). The fact the chip shortages still continue seems to elude our new political masters ....... However, the US Military DOES STILL CLEARLY INSIST that more defend-able "in USA" production facilities be opened ASAP due to China's military flying dive bomber jets right over the TSMC production facilities just to INTENTIONALLY point out how completely vulnerable our chipset supply line really is ........ =================================================== AMD will be using a chiplet approach at 3nm for both video and for CPU cores. 5nm will likely still be APU based for the laptop stuff with 3nm being so low in power consumption that there likely won't be much real difference between laptop and desktop processor products except core counts, graphics levels, etc. This is pretty much business as normal in the phone world. TSMC has bluntly said this week that Intel is not currently considered a long term TSMC partner, but barely rates as a second level "occasional customer" in TSMC's eyes. Face it, Intel wants time on the older lithography systems right now -- not the brand new ones. Intel is saying this isn't true and that they do have allocations, but Intel's marketing department has a long long history of lying to stockholders and to financial reviewers and to all the technical press. ::) :P who knows what is real, it is "Intel Inside" that is talking after all ......... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/06/21 at 21:14:16 https://www.tomshardware.com/news/intel-fires-back-at-apple-m1-processors-with-benchmarks This is a current example of false Intel marketing. Intel cannot beat Apple's M1 chipset unless they wring in a special Intel processor equipped Laptop in the hands of Principled Technologies (you remember them, the little cheat firm Intel uses to lie to people about processor performance). Principled Technologies applied all the dirty tricks we remember from the Tiger Lake introduction and it appears they are working up the same set of BS for the Rocket Lake introduction. There are Many Many Many things wrong with Principled Technologies benchmarking and with Intel for using this data in advertising the Tiger Lake and Rocket Lake processor families. Most egregious is the blatant swapping off of hardware (test to test) in order to show a false result --- this is clearly intentional fraudulent false testing. Interestingly, in the configurations document at the end of the slides, Intel shows that it switched to a MacBook Pro with 8GB of RAM, rather than the 16GB model it tested for performance. In battery life, Intel switched to an Intel Core i7-1165G7 notebook, the Acer Swift 5, rather than sticking with the Core i7-1185G7 in the whitebook it used for performance testing. It also tested a MacBook Air. They ran Netflix streams and tabs and found the MacBook Air came ahead with a six-minute difference. Intel didn't list battery life for the MacBook Pro. In our tests, that beat Intel PCs by hours. Folks, Intel has not changed one wit internally with their new CEO in place. Intel's first reaction is to LIE and CHEAT ....... if you see Principled Technology mentioned (or its predecessor, Trusted Technology) you KNOW what Intel is up to ----- they need a cut out to blame all the false testing upon. Just like with Tiger Lake introduction, Intel will now flood the news channels with the false testing results and will have a whole herd of shills out in the review world pushing this illicit testing bullshite on into the minds of the purchasing public. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/07/21 at 08:13:57 https://www.google.com/search?q=intel+cheats+on+benchmarks&oq=intel+cheats+on+benchmarks&aqs=chrome..69i57.18734j1j15&sourceid=chrome&ie=UTF-8 https://www.extremetech.com/computing/278661-principled-technologies-responds-to-intel-whitepaper-controversy-we-test-performance-claims https://youtu.be/qzshhrIj2EY This is a YouTube video of a physical visit to Principled Technology in Durham NC (which is a marketing firm, not a comparative review test house). Some of the responses to Intel's intentionally using Principled Technology to test for them leave the viewer quite frustrated. And then there is Principled Technology's somewhat lame responses to getting called out on the previous mess they participated in at Intel's request. PT basically did what Intel asked them to do, and PT is continuing to deflect most of the heat from Intel's poor real performance by obfuscating the heck out of the testing that was done. Does PT even have a lab that is able to do this testing? And where did all the hardware variations reside and how did they get confused into the results? With all the PT testing folks supposedly available downstairs SOLID ANSWERS could have been gotten with a quick phone call, but this was not what happened at all ...... Instead, you are skillfully left feeling like poor little PT is jest getting picked on by people who just want to catch Intel's hand in the cookie jar (again). And yes, that actually is the case in actual reality, so I guess PT is worth the money they get paid for insulating Intel from the results of their tilted test program. It is clear that the person spoken with is very carefully "not technically proficient" to the degree that is needed to answer any questions. This does not remove the testing bias and the grotesquely inappropriate switching out of hardware that was done --- and no I do not believe PT DOES NOT KNOW how all that happened. Fact remains, Intel is using a fairly slick acting Marketing Firm to "originate" the independent support documentation for their Tiger Lake and Rocket Lake product claims. Intel sux ....... to represent this shyster firm as a "technical review firm" for complex computer processor testing is totally ludicrous. Go here and read all about Principled Technologies and what their self-given mission statement is. I find it enthralling that https://in.linkedin.com/company/principled-technologies this guy, https://in.linkedin.com/in/seanpatrickkelley?trk=org-employees_profile-result-card_result-card alias Sean Patrick Kelly, the head dog, calls himself (in this order) sales leader - marketer - storyteller. No wonder he can't answer any technical questions ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/08/21 at 02:17:27 https://www.tomshardware.com/features/amd-vs-intel-cpus Back in first quarter 2021 reality land, Intel has bumped AMD once again ----- and guess what, the AMD counter wave took a big 4 days to come back in and totally flood the Intel victory sandcastle. Hump for bump, tit for tat, swap for bap ...... the only victor here is the consumer who is getting better and better chipsets all the time. Intel is positively straining itself to find new ways to cheat and AMD keeps coming out with MORE REAL IMPROVEMENTS to overwhelm Intel's BS results. THIS IS NOT 5nm GENOA, see next post below this one for Genoa specs. NOTE PLEASE: "Substrate scarcity" has driven chipset prices on the lowered production levels right on up into the clouds, however ............. http://https://cdn.mos.cms.futurecdn.net/YqfpCwKBTtW3y5Xh2AKkw5-970-80.png.we Has anyone else noticed that the Ryzen 5 chipset in each generation rates the same as the more populated Ryzen core counts? Has Tom's Hardware Geomean testing driven the results right on down to the chiplet level? Cinebench 23 testing still shows a gradient of results which I think benchmarks need to provide. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/09/21 at 05:35:35 https://wccftech.com/amd-5nm-next-generation-zen-4-ryzen-epyc-cpus-to-feature-over-25-ipc-increase/ https://www.notebookcheck.net/Zen-4-rumor-bulldozer-smashes-Zen-3-with-alleged-25-IPC-uplift-40-total-gains-and-potential-delivery-of-5-GHz-on-all-cores-while-AMD-s-Zen-5-will-be-even-more-spectacular.519409.0.html AMD’s 5nm Next-Generation Zen 4 Ryzen & EPYC CPUs Rumored To Feature Over 25% IPC Increase, 40% Overall Performance Boost Over Zen 3 http://https://cdn.wccftech.com/wp-content/uploads/2020/04/El-Captain-Supercomputer_AMD-EPYC-Genoa-Zen-4-Radeon-Instinct-Next-Gen-GPU-Accelerator_3-1030x445.jpg The overall performance gain for the Genoa CPUs could very well start around 40% considering that the CPUs are going to feature more than just a core architecture upgrade. There will be increased clock speeds and also key improvements to the IMC along with support for faster DDR5 DIMMs and an improved interconnect fabric that will communicate with the several cores. The Zen 4 CPUs are stated to offer an increase in core count which was hinted by AMD's CEO, Dr. Lisa Su, in a previous interview. The 4th Gen EPYC Genoa CPUs are expected to feature up to 96 cores though we could end up with an even higher number if AMD really wants to push the boundaries in the server CPU segment. Intel, crank up on your lies right smartly now boys, 40+% is a whole lot of BS that you have to shift very quickly knowing full well that another improved AMD reality wave will actually arrive before your bogus exaggerations and lies can all become "gospel" as pushed out by your army of shills. AMD has wafers of 5nm chiplets on hand now from both TSMC and from Samsung and is currently testing out the first assembled AMD chipset layouts ...... and based upon the AMD socket size remaining the same and the chiplets getting so much smaller, AMD can tuck 2-4 more chiplets into each processor layout (which 2-4 depends on the current chiplet count and layout pattern of course). Each chiplet has 40% more throughput and we are told that 5ghz chiplet speeds will happen in this set of generations as well as PCIe 5.0 support and even greater increased bandwidths which will be hitting with the mainframe roll ins. As well as 14 layers per chiplet for more actual on-chip memory and for more built in Xilinx functions. AMD has several serious generations of improvements to put into their 5nm chiplets, so LIE REALLY BIG Intel, lie really really big. When I first heard that the proposed Ryzen 3 outperformed the current Core i9 I was somewhat unbelieving, but now I see how this can be possible as Ryzen 5 certainly does outperform Core i9 and has for the last 2 generational exchanges. ================================================== http://https://www.notebookcheck.net/fileadmin/_processed_/e/8/csm_zen_ipc_excavator_2_1c81f3d59a.jpg https://www.notebookcheck.net/Zen-4-rumor-bulldozer-smashes-Zen-3-with-alleged-25-IPC-uplift-40-total-gains-and-potential-delivery-of-5-GHz-on-all-cores-while-AMD-s-Zen-5-will-be-even-more-spectacular.519409.0.html Goes over the same basic ground as the wccftech.com leak listed above, but offers some 3rd source confirmations of some of the key details. That is 3 sources in confirmation, so it is pretty solid I would say. Plus AMD's habit of having the actual factual before they talk about it ...... and then shipping the same on time. SPEAKING OF NOT ON TIME ...... Intel has added another full year or so on to the dubious arrival time of Intel 7nm, which puts it a year out past when TSMC and Samsung will be building stuff for Intel in a major fashion according to Intel. Yep, all the way out there in year 2023-24. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/11/21 at 05:49:27 https://prog.world/amd-epyc-genoa-on-zen-4-architecture-as-early-as-2022-and-rumors-about-smt4/ NOW LET'S THNK MAINFRAME FOR A BIT, without getting it confused with consumer or gaming in our heads. This is useful to us as it predicts more exactly what AMD will be doing and when it will be doing it as the tech needs to be built in at the chiplet level to be there for mainframe uses when the time comes. AMD EPYC GENOA on Zen 4 architecture as early as late 2021 and new development rumors about SMT4 multi-treading and the AM5 processor socket AMD has just announced the custom Ryzen series processors on its new Zen 3 architecture and is still preparing the corresponding server announcements, when information about the next generation of the company’s server processors – AMD EPYC GENOA on Zen 4 – appeared on the network. http://https://i1.wp.com/habrastorage.org/webt/xp/z1/zo/xpz1zo2dyddreqiqxdbymahaex8.png?w=1160&ssl=1 We are talking not only about processors with up to 96 cores, but also about new generation sockets AM5, TR5 (HEDT platform), as well as SP5 and SP6 (server platform). It also talks about support for PCI-Express 5.0 and DDR5 memory. While the information circulates on the network at the level of rumors, but given the way AMD’s consumer segment is developing, their attempt to regain ground in the server market is a matter of time. So, EPYC MILAN on Zen 3, according to inside information, will be at least 20% more productive than the previous generation of AMD processors of the Zen 2 ROME architecture. Now let's carefully shift our minds to a mainframe point of view concerning these things, and REMEMBER please that all of these advancements will have to be built into the same base chiplets that get used everywhere in the AMD product line. If we talk about AM4 vs AM5 sockets, then it becomes clear from rumors that AMD has squeezed everything it could out of the AM4 platform as such AMD has made a huge gift to consumers without updating the socket for Ryzen 5xxx while ensuring backward compatibility of new 5xxx desktop processors with an existing socket. Here you can recall the endless changes to sockets at Intel, of which at least three have been released over the past four years: LGA 1151, LGA 1151 v2 and LGA 1200. In 2022, however, A4 will be replaced by the A5 socket and, hopefully, the A5 will live as long as the A4 did. We know that the large server sockets will get updated: we will switch from Socket SP4 and Socket SP4r2 to SP5 and SP6. Most likely, both sockets will be released at the same time and will be suitable for the same generation of EPYC GENOA with the same difference as the SP4 and SP4r2 sockets: the former are designed for single-threaded processors, and the latter are for dual-threaded EPYC ROME processors. If we assume that AMD still implements SMT4, then SP4 will work, respectively, with single-threaded and dual-threaded processors, and SP4r2 – with four-threaded models. But it’s worth returning to the topic of SMT. A possible path to further AMD's leadership is to increase the number of threads per core…... We are talking about the simultaneous processing of four threads, instead of the standard two threads per core. It should be noted that SMT4 technology is almost guaranteed not to be present in EPYC MILAN processors, but will likely be present in the follow on EPYC GENOA processors in 2022. You need to understand that in the home use segment of the market SMT4 is a dubious feature that will only tend to increase latency and create minor timing losses out of the blue while the processor decides on which of one out of these many threads to transfer the next task, out of the dozens and hundreds of threads that various applications will create in real time. Deciding takes time, sometimes more time than is saved by using the extra threads. However, in the case of the pure mainframe processor segment, there we are dealing with more simplified more monolithic systems, which, more often than not, are sharpened for working with some relatively few heavy software complexes. This is where SMT4 can do very well, especially in mainframe computing. At the same time, SMT4 and even SMT8 are actually a set of ancient systems that were used starting over twenty years ago. The development of multi-threading peaked in the early 2010s, when IBM still had its own production of servers for business. The IBM POWER8 Core had variable multithreading, ranging from one core to one thread to eight threads per logical processor core. The official POWER8 Core frequencies per core range from 2.5 GHz to 5 GHz. At the same time, IBM servers on POWER8 also had 16 sockets SMP (symmetric multiprocessing) – which made it possible even then to combine a dozen servers into a computing cluster. It is worth noting that IBM servers were a very specific and narrow solution for large corporate business and scientific computing. Actually, with the growth of AWS and Azure, they were squeezed out of this segment and IBM Power S822LC became the last product of the company in this line. I must say that now, Intel’s Xeon processors, which have practically captured the server market, also do not work with SMT4 mode. If we are talking about processors for “science” – that is, about monstrous solutions for 32-72 cores of the Phi series, for example, Intel Xeon Phi Processor 7295 with 72 cores and a cost of ~ $ 6200 at the time of release, then we do not have multithreading at all … According to the official specification, this processor has 72 cores and 72 threads. The more popular Intel Xeon E operate in SMT2 mode – two threads per core. This applies to almost all popular Intel server processors released since 2013, starting with the E5-V2 series. To give a concrete example, two threads have already been in the extremely popular workhorse in the form of the Intel Xeon E5-2680V2 processor, which is still actively used. If the forecasts and data from insiders are confirmed, then in the next five years we can expect a partial redistribution of the server market. AMD is actively developing in the desktop direction, now it is starting to put pressure on the server segment. Intel plans to announce server processors this spring, but it is not a fact that they will be much more powerful than the new MILAN, and even more so the upcoming GENOA. “Blues” are still experiencing significant difficulties with lithography and cannot really go below the 14 nm process technology, somehow launching the production of 10 nm laptop processors. It's not hard to figure it out, Samsung builds it for them right now because Intel can't build it at all ...... Why pay attention to these mainframe leakers ???? Because mainframe people don't care about PC gaming or general home use functionality and they simply don't see their mainframe world as being the feature predictor of the entire AMD stack. Because of their long lead times for decision making, AMD gives future info out to the mainframe folks first, giving it to them far enough out for them to be able to use it in their longer extended planning horizons. Is this information reliable? Yes, the info is VERY reliable --- lie to the mainframe elephants just once and they will NEVER forget it, so AMD is careful to only tell them the actual factual. Also please note that the integration of Xilinx into AMD chipsets would also take place in this same time frame, but this is not being revealed or discussed at all by AMD until the sale has full approval by the FTC and it is actually a done deal. Being a simple stock swap arrangement inside the USA, involvement by other countries is not required for the ARM Xilinx arrangement. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/14/21 at 01:20:26 https://www.google.com/search?q=qualcomm+protests+ARM+purchase+by+Nvidia&oq=qualcomm+protests+ARM+purchase+by+Nvidia&aqs=chrome..69i57j33i10i160.18611j0j7&sourceid=chrome&ie=UTF-8 https://finance.yahoo.com/news/google-microsoft-qualcomm-protest-nvidia-183045665.html (Bloomberg) -- Some of the world’s largest technology companies are complaining to U.S. antitrust regulators about Nvidia Corp.’s acquisition of Arm Ltd. because the deal will harm competition in an area of the industry that is vital to their businesses. Alphabet Inc.’s Google, Microsoft Corp. and Qualcomm Inc. are among companies worried about the $40 billion deal and are urging antitrust officials to intervene, said people familiar with the process who asked not to be identified because they weren’t authorized to speak publicly. At least one of the companies wants the deal killed. Nvidia shares fell as much as 3.1% in New York trading on Friday. The acquisition would give Nvidia control over a critical supplier that licenses essential chip technology to the likes of Apple Inc., Intel Corp., Samsung Electronics Co., Amazon.com Inc. and China’s Huawei Technologies Co. U.K.-based Arm is known as the Switzerland of the industry because it licenses chip designs and related software code to all comers, rather than competing against semiconductor companies. The concern is that if Nvidia owns Arm, it could limit rivals’ access to the technology or raise the cost of access. Nvidia has argued that the purchase price alone means it has no incentive to mess with that neutrality but some rivals and Arm customers are unconvinced. “As we proceed through the review process, we’re confident that both regulators and customers will see the benefits of our plan to continue Arm’s open licensing model and ensure a transparent, collaborative relationship with Arm’s licensees,” an Nvidia spokesperson said in a statement. “Our vision for Arm will help all Arm licensees grow their businesses and expand into new markets.” Google, Microsoft, Qualcomm and Arm declined to comment. CNBC reported Qualcomm’s objections earlier. Before the deal can close, Nvidia must get through a long review process by antitrust officials in the U.S., U.K., European Union and China. Government agencies globally are in the process of reaching out to those they believe may be affected by the transaction. A groundswell of opposition from large tech companies may make it difficult to win approval, delay the process or force concessions that change the value of Arm to Nvidia. This is also a risk for SoftBank Group Corp., the current owner of Arm. The Japanese conglomerate has been trying to sell some assets to pay down debt and buy back stock. In the U.S., the deal is under review by the Federal Trade Commission, which has opened an in-depth investigation of the merger and has sent information demands to third parties, according to a person familiar with the matter. The FTC declined to comment. OK, why does this matter? All of the major computer companies have now lodged protests over the Nvidia purchase of ARM Holdings. Jensen "Leatherman" Huang's purchase of ARM will inevitably affect a whole host of major players as Huang has always been an aggressive rule-bending player who seeks to dominate and manipulate entire markets. Giving him ARM Holdings to use as a "move the world pivot point" and "a long enough lever" would be a VERY BAD idea, and all the various computer world players are clearly saying so in concert right now. And not just to their own regulatory bodies, but to each one of the 4 major country regulatory bodies, each regulatory body in turn being told clearly that this is a bad idea and exactly why it is bad for their country in particular. With this large amount of organized resistance, the chances of Huang being able to make his deal are lessened greatly. Plus, in the USA the structure of the FTC itself has changed drastically with the change in administrations. In the U.S., the deal is under review by the Federal Trade Commission, which has opened an in-depth investigation of the merger and has sent information demands to third parties, according to a person familiar with the matter. The FTC declined to comment. The changing leadership of the FTC could make winning approval tougher for Nvidia. The commission is generally split 2-2 along party lines at the moment, with Democratic commissioner Rebecca Kelly Slaughter holding the acting chair position. Power will shift to the Democrats when U.S. President Joe Biden picks two candidates to fill an open seat and the seat held by Commissioner Rohit Chopra, who has been nominated to take over the Consumer Financial Protection Bureau. Deals like Nvidia’s acquisition of Arm, known as vertical mergers, are typically seen as less worrisome in the eyes of antitrust enforcers because the companies don’t compete head to head. But that view has come under fire from advocates of more aggressive antitrust enforcement who say regulators have downplayed the competitive harm from such deals. Slaughter’s elevation signals a tougher approval process for vertical deals. Before taking over the agency, Slaughter criticized new guidelines issued last year by the FTC and the Justice Department outlining how the agencies would evaluate vertical deals. She said the guidelines overemphasize the potential benefits of such mergers and are “inexplicably mute” about the harms. In December, Slaughter and Chopra said companies should no longer rely on the Trump administration's guidelines as an indication of how the FTC will police vertical deals. “Moving forward, we need to aggressively enforce against the harms of vertical mergers,” they wrote. “We look forward to turning the page on the era of lax oversight and to beginning to investigate, analyze, and enforce the antitrust laws against vertical mergers with vigor.” Hwang's deal sounds like an excellent whipping boy for the Democrats to use to show the world that they are large and in charge again ....... These same protest techniques may be used against the AMD Xilinx merger as Intel for one would leave no stone unturned in their attempts to hamstring AMD. The elevation of the major computer chip businesses to NATIONAL SECURITY status means their health and well being will get politically fiddled with more and more as the months go by ..... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/17/21 at 01:14:04 https://www.tomshardware.com/news/samsung-hbm2-hbm-pim-memory-tflops https://www.zdnet.com/article/samsung-develops-high-bandwidth-memory-with-integrated-ai-processing/ As Intel struggles to regain relevance, Samsung and AMD-Xilinx and several others strike out in new directions that will render traditional Intel x86 completely nonsensical. Today, Samsung announced that its new HBM2-based memory has an integrated AI processor that can push out (up to) 1.2 TFLOPS of embedded computing power per chip, allowing the memory chip itself to perform operations that are usually reserved for CPUs, GPUs, ASICs, or FPGAs. The new HBM-PIM (processing-in-memory) chips inject an AI engine inside each memory bank, thus offloading processing operations to the HBM itself. The new class of memory is designed to alleviate the burden of moving data between memory and processors, which is often more expensive in terms of power consumption and time than the actual compute operations. Samsung says that, when applied to its existing HBM2 Aquabolt memory, the tech can deliver twice the system performance while reducing energy consumption by more than 70%. The company also claims that the new memory doesn't require any software or hardware changes (including to the memory controllers), thus enabling a faster time to market for early adopters. Samsung says the memory is already under trials in AI accelerators with leading AI solutions providers. The company expects all validations to be completed in the first half of this year, marking a speedy path to market. OK, think of all the flexible processing techniques that are already in use. Now think about all the additional layers in your 14 layer 5nm processor stacks, plus all the GREATLY increased flat space per layer that will exist at the 20 layer 3nm stacked lithography. Now partner up Samsung with AMD-Xilinx and use all that flat 3nm 20 layer stack space to put a very strong AI processor actually inside your existing memory lay outs ....... This is sensible stuff that is already in trials right now, it is REAL STUFF that is ready to roll out right now. Why have rack after rack of Nvidia video cards burning up the watts trying to do your AI stuff when you can do this right inside your systems memory at far far less energy cost? This changes the game up quite a bit for those innovative and flexible enough to go for it, right now. http://https://www.extremetech.com/wp-content/uploads/2012/04/scandia1-640x508.jpg This is a 1990 ASCI Red, the very first Intel x86 supercomputer to go over a teraflop of processing power. Each stacked chipset of the new Samsung HBM2-based memory has 1.2 of these babies functionally tucked inside it, except that the new Samsung AI chipset is a whole lot quicker off the mark and comes with software and applications that can actually USE this level of computing power. Change, she comes ...... ::) Notably, the paper and slides above refer to the tech as Function-In Memory DRAM (FIMDRAM), but that was an internal codename for the technology that now carries the HBM-PIM brand name. Samsung's examples are based on a 20nm prototype chip that achieves 2.4 Gbps of throughput per pin without increasing power consumption. The South Korean tech giant explained that this extreme progress was possible as the installation of AI engines inside each memory bank maximizes parallel processing while minimizing data movement. By providing this improved performance, Samsung said it expects the new chip to accelerate large scale processing in data centers, high-performance computing systems, and AI-enabled mobile applications. The 3nm version of this tech next year will be a lot more capable, of course. And consume proportionally far far less power as well. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/17/21 at 21:00:04 https://www.geekwire.com/2021/chromebooks-outsold-macs-worldwide-2020-cutting-windows-market-share/ http://https://cdn.geekwire.com/wp-content/uploads/2019/08/dell-latitude-5300-2-in-1-and-5400-chromebook-enterprise_2x-e1566839929232-630x296.png This is a pair of Dell Business class Chromebooks specifically designed to replace Windows Laptops in a heavy data entry environment. Chromebooks outsold Macs worldwide in 2020, cutting into Windows market share New numbers show 2020 was the first year that Chromebooks outsold Macs, posting impressive market share gains at the expense of Windows. Computers powered by Google’s Chrome OS have outsold Apple’s computers in individual quarters before, but 2020 was the first full year that Chrome OS took second place. Microsoft’s Windows still retained majority market share, but also took a big hit as both Chrome OS and macOS gained share. The milestone is based on numbers provided by IDC, which doesn’t typically break out sales based on device operating system. But when we went looking to see how the pandemic may have impacted the PC market, IDC analyst Mike Shirer confirmed the findings to GeekWire. (We also contacted Gartner but that firm does not include Chromebooks in its traditional PC market results.) This is a big win for Google and a warning for both Apple and Microsoft. It also signals to app and game developers that Chrome OS can no longer be ignored. Frankly, any business that provides a product or service over the internet should be setting aside resources to ensure the Chrome OS experience is comparable to Windows and macOS. Breaking down the numbers for all of 2020 http://https://cdn.geekwire.com/wp-content/uploads/2021/02/marketshare.png Q3 2020 and Q4 2020 confirmed the trend: Windows dropped further to 78.9% for Q3 and then 76.7% for Q4; macOS grew to 8.4% for Q3 and then fell back to 7.7% for Q4, while Chrome OS had 11.5% for Q3 and then 14.4% share for Q4. The Q4 results are particularly notable as the fourth quarter tends to be the biggest for PC shipments. While macOS widened its lead over Chrome OS in Q4 2019, Chrome OS came roaring back in Q4 2020. I was struck last quarter by how quickly and firmly Samsung and AMD-Xlinx bonded together in tech sharing arrangements over video graphics tech and "other stuff". I would say that Intel's secretive use of AI to make up for their lagging chipset technology is no longer unique nor really particularly secret at this particular point in time ....... MediaTek in particular is treating Chromebooks seriously, designing some very modern "PC class" chipsets for that particular use. I look for Chromebooks to pick up some serious ARM based processing horsepower in 2021-22 especially if they are using the Samsung memory processing tricks and the various other new XPU AI processing techniques. https://youtu.be/QBL07SKYC7I it is a YouTube about the new Mediatek chips being functionally 5-10 times more powerful than the Chromebook chips what we are used to seeing from Rockchip and from Mediatek previously. Also note: Samsung is bigger than TSMC when counting wafer output numbers, and Samsung is now growing faster than TSMC is too. Watch Samsung build a BIG new plant in Austin Texas this year. A new term for everyone to know and use, XPU instead of CPU It means a processor that is pumped up by AI (done openly using known open standards). poor Intel, poor Wintel too. If you add together the market share for Chromebook and for Apple M1 laptops that same number equals the market share loss for both Intel and for Microsoft. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/21/21 at 17:02:27 New Chromebook OS will have more power using same chipsets, but 5x more power using special built XPU chipsets that are on the way for second half 2021. Chromebooks have all the newest Google drivers for the full AI capabilities of the chipsets they are built with. This includes the built in AI and graphics accelerators in the XPU or in the memory itself. You can use Steam or Google gaming to get your gaming on with your new more powerful Chromebook ....... Chromebooks can run Chrome apps, Android apps and Linux Apps along with any web based app set out there (Microsoft is included in this mix for free). Chromebooks always cost less (simple, Chrome OS requires less hardware to run well). ::) |
|
Title: Re: AMD & others --- post AMD dominance in 2020 Post by Eegore on 02/22/21 at 20:27:24 "Chromebooks can run Chrome apps, Android apps and Linux Apps along with any web based app set out there (Microsoft is included in this mix for free)." Chromebooks remind me of the old video game wars of the late 90's. Nintendo and Sega fought for content superiority and then one day PlayStation came in and said "We don't care "who" make the games, we just want them in our library." They just wanted a working console with lots of games, didn't care about exclusivity, Chromebooks seem similar, they just want to focus on the user and what the user wants, not some proprietary software or functionality. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/22/21 at 20:46:34 Yep, Eegore Chromebooks run EVERYTHING other than formal Microsoft products. And it is funny, the newer Chromebooks can run MS products off the web and can run full package installs of the MS Office for Chrome variants, too, but nobody wants to pay the Mickey Tax for that sort of action. Chromebook people are generally already really good with Google Docs and with Libre Office Suite, both which are excellent FREE alternatives to Microsoft Word, Excel, etc. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/22/21 at 21:03:32 https://arstechnica.com/gadgets/2019/11/intel-core-i9-10980xe-a-step-forward-for-ai-a-step-back-for-everything-else/ OK, you got to go look at this one since most of the graphs won't do cut and paste. What does it actually mean and why does it matter? Threadripper absolutely kills the Intel product in all normal tests, slaughters it by a factor of two or more until the first AI based test (from Intel of course) is applied and suddenly the inverse is true. Intel then kills the Threadripper just as badly on the one (1) scenario that Intel has set up for AI augmented benchmarking. http://https://cdn.arstechnica.net/wp-content/uploads/2019/11/TR3970x.004-640x480.png This is one hell of a 4x+ shift in performance all of which is due to AI and none of which is due to the Intel processor per se. What does it all mean? In general terms we are on the cusp of the AI Accelerator Age. Look to see AMD-Xilinx-Samsung-IBM alliance start to throw up combined tech package standards that slay whatever Intel puts together, then Intel will strike back with a new version of their stuff and then it will devolve into tit for tat all over again. The AI Age will play out for serious first in the bigger scale mainframe equipment, but AI Age will make it down to PC class very quickly as the Hockey Stick Phone Boys play in this ballpark already in Cell Phone Space. ::) Samsung is signalling it will push this from both the PC and the Phone perspectives. Apple will be there in both worlds as they roll into their ARM PC chipsets. AMD will play there too, on the PC side just as soon as the Xilinx purchase is considered "confirmed". Intel is there already, but Intel only plays in Benchmark Space at the moment, nothing is real for productivity (apart from resizing some graphics images that is). Tom's Hardware, Arstechnica, Anandtech and others are now working on AI testing methodologies ---- all the major software suites are now being changed so that applicable AI calls are part of the package. Ditto for the basic OS operating calls, AI is going in everywhere it can apply because AI offers a whole lot of bang for the buck. ::) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/24/21 at 20:12:43 https://liliputing.com/2021/02/samsung-could-launch-a-windows-pc-with-exynos-cpu-and-amd-graphics-this-year.html Samsung could launch a Windows PC with Exynos CPU and AMD graphics this year http://https://liliputing.com/wp-content/uploads/2021/02/samsung-galaxy-book-s_03-700x418.jpg ZDNet says the processor, which might be known as the Exynos 2200, should be ready in third quarter of 2021, which is when Samsung plans to debut the processor in its new Windows on ARM computer, which will most likely be a thin and light laptop similar to last year’s Samsung Galaxy Book S. It’s unclear how Samsung’s new chip will stack up against Qualcomm’s Snapdragon 8cx Gen 2 processor, which is currently the most powerful chip used in Windows 10 on ARM computers. But hopefully a bit of competition in this space will drive Qualcomm and other chip makers to step up their game in terms of performance. =================================================== Intel's Rocket Lake midrange chipset is out and has now been benchmarked against AMD's Ryzen 7 from last year. Intel was boning hard at getting back at least one gaming crown, and they did so at a simply astronomical cost in energy & efficiency and by using all their multiple AI tricks. Using all Intel's AI tricks and running at a very very hot very high current draw (hyperclocking) rate, Intel managed to eek out a 2% win over AMD's bog standard no tricks no bullshite processor benchmark rating. 2% isn't much, and AMD could counter this with a very slight processor speed bump of their own lower lithography 7nm chipsets vs the Intel 10nm competition. But AMD won't do this because the next shift they plan in their own AMD processors (to 5nm) wipes out Intel's tiny 2% lead quietly and naturally as part of AMD's 15%+ IPC improvement. Plus, there are over 6 AMD chipset overclocking utilities out there now (3 are open source and completely free to use) that allow those users who want to go tit for tat with Intel's hyperclocking to do so freely on their own. AMD simply isn't going to play Intel's mickey mouse tit for tat games with "over running their voltages and their clock speeds" ........ ================================================== We are looking for the substrate based shortages to last for the rest of this year, possibly into 2022 as well. You are likely to see more severe AMD shortages showing up as everyone wants the AMD chipsets right now and Intel is always trying to get closer to competitive so they can sell their products too. Hard fact is, that with both of our USA suppliers running as good as they can right now America is going to continue to fall short on available products ---- and even more Near Eastern ARM based Chromebooks will be sold in the USA to fill in the blanks. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/01/21 at 19:03:12 RUMORS AND LEAKS Very briefly, AMD mainframe chipsets are being put together now for the next AMD generation using 5nm chiplets and a new center communications chiplet. Roughly you have a third more to twice as many more cores per existing processor type up and down the product stack. You have twice to three times the internal cache memory amounts for the 3 existing cache levels, and you can share the new last largest cache between all the processor threads through roughly twice as many much faster data pipelines. PCI Express 5.0 spec will now be supported for the first time (but use is somewhat limited right now as the 5.0 PCI Express standard is brand new right not) as well as the existing PCI Express 4.0. All of these benefits dwell inside the standard 5nm chiplets that will get used all over AMD's product lines. Mainframe chipsets change relatively more slowly and remain in use for a longer time compared to PC stuff, so you can expect these new mainframe features to stick around for a while. And yes, some of these rich new features are built into all the 5nm chiplets now, but are not initialized yet by motherboard BIOS nor by the OS software currently being used. Layouts seem to run six groupings of chiplets instead of four groupings of chiplets as was done in the past as the 5nm chiplets being that much smaller. Used to be grouped in 2 pairs of 2 (ie. 4) are now grouped in 6-8 per group with no subgroups of two splitting anything up. TSMC is busy raising 800 million each from Texas and Arkansas to fund the local part of their new TSMC manufacturing plants that the US government has asked TSMC to build. This brings the TSMC total new American plant count up to 6 plants with at least one being a TSMC megaplant. We expect some of these plants will go in at 3nm lithography as the timelines for these plants add up with what ASML is actually building for scanner lines right now (5nm scanners are all built out at this point in time and ASML has rolled over to 3nm scanners on all future orders). China is not being allowed to buy these modern scanner lines (Entity listed, remember) but they are sneakily ordering some older more obsolete lines from the Japanese used equipment dealers with obsolete being 7nm and older more precisely. Biden may act to shut this down shortly. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/02/21 at 10:03:16 https://videocardz.com/newz/amd-epyc-genoa-zen4-to-feature-96-cores-across-12-chiplets-12-channel-ddr5-memory-and-sp5-lga6096-socket https://videocardz.com/newz/amd-zen4-core-in-epyc-genoa-may-support-avx3-512-instructions http://https://cdn.videocardz.com/1/2021/02/AMD-EPYC-Zen4-Genoa-Mockup-768x432.jpg Confirmations of the latest rumors from more detailed sources about AMD 96 core Genoa. In the last 48 hours, numerous details on AMD Zen4 in EPYC 7xx4 series codenamed “Genoa” have leaked and 24 hours later have been confirmed. We have learned that the new server processors will be 96 core and support 12 channel DDR5 memory, but we also learned it will have PCI Gen5 support. The internal configuration of the chiplets has also been leaked. AMD Genoa will require a new SP5 socket. This socket will support CPUs with the LGA6096 package measuring 72×75.4 mm, according to another leak from Patrick Schur. The package is not a perfect square, but not as rectangular as SP3-based EPYC processors. The new package will have a lot more room for future chiplet designs, with the first being 12 chiplets plus 1 I/O die. The first generation SP5 EPYC will be limited to 96 cores though, according to a leak from ExecutableFix. Today a part of a new slide deck has allegedly leaked on Chiphell. The slide appears to list all features of the new Zen4 cores, confirming it would have more than 64 cores and support 2 threads per core. It was previously rumored that Genoa might have 4 threads per core, but that was quickly debunked. More importantly, though, the slide would suggest that Zen4 supports AVX3-512 and BFLOAT16 instructions. Both are extensively used in deep learning computations. The AVX 512 support on Intel CPUs and accelerators one of the key advantages over the AMD EPYC series. If this leak is true, then this is about to change. ExecutableFix, who yesterday leaked the Zen4 details, has already confirmed that AVX3-512 support is indeed coming to the EPYC series. AMD EPYC “Genoa” based on Zen4 core design is now expected to launch in 2022. AMD is yet to unveil its Zen3 based Milan series, a successor to Zen2 based Rome. We have confirmed that Apple is up and running on 3nm TSMC ---- this means AMD now rolls into the use of the 5nm lines that are now being freed up by Apple. Thus the sudden waves of AMD 5nm chiplets and new 5nm AMD processors hitting Rumorville ...... ::) poor Intel, still 2-3 generations behind ........ |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/04/21 at 08:22:44 2021 will be a year of shortages and limited volume introductions. Apple is already full production on at 5nm and Apple has begun rolling over to 3nm as we speak. AMD is just now rolling out their first bulk production wafers of 5nm chiplets and connection center chiplets at TSMC (as per contract) but this year is going to be an allocation nightmare with AMD being on 7nm and on 5nm shifting over one product line at a time. This will tend to delay shifting over the cheaper, lower end AMD products while all the carefully sorted 5nm chiplets get allocated out among the larger, more profitable product lines. 3nm will also start AMD chiplet production sometimes late this year, at which time the total chip shortage picture should ease up some, but there will not be enough 5nm or 3nm to make all the chiplets AMD needs, not right at first. This issue springs from American Automotive component makers not allocating what they needed in time, so with carmakers are now playing politics with their plant shutdowns to get a relatively forceful political push to get the Automotive chipset needs rung in "after the fact" ---- putting everything else off kilter for another degree or three. So 2022 for AMD may be an even more mixed 7nm, 5nm and 3nm bag until enough production capacity exists to roll all of AMD's volume to the current best lithography level. Intel does not seem to have any TSMC allocations set up at this time. 3nm is the earliest that Intel seems to have a foot in the water. Intel has a longstanding habit of not announcing when they get a competitor to run their chipsets for them (corporate image and all that I suspect) so not seeing Intel as a TSMC customer may not be conveying any real information to us. We do know the new Intel video chipsets are run at either Samsung or at TSMC, but which supplier is doing that business right now is a mystery ....... (last time it was known it was Samsung) ::) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/07/21 at 12:09:59 https://www.notebookcheck.net/Ryzen-7000-rumors-start-flying-as-Phoenix-is-revealed-as-AMD-s-likely-Zen-4-successor-to-Rembrandt.523163.0.html MORE RUMORS FLYING, THIS TIME THEY ARE MOBILE APU RUMORS http://https://www.notebookcheck.net/fileadmin/_processed_/e/9/csm_AMD_Ryzen_7000_Phoenix_drd_297dde7a7d.jpg The post reveals that Phoenix, which has been vaguely hinted at before in an alleged roadmap that simply stated “P******” for the future of AMD’s mobile parts, has the CPUID A70F00 and more tellingly will use the FP8 socket. This shows that Phoenix is a mobile processor, as the current Cezanne Ryzen 5000 mobile APUs utilize FP6. This is where a lot of rumors start flying, such as Rembrandt, the Ryzen 6000 mobile APU family based on Zen3+ microarchitecture, being the only chip series to use an FP7 socket. Furthermore, it’s also expected that Phoenix will offer DDR5 support, which is also supposedly coming with Rembrandt anyway, and that it will use the 5nm manufacturing process. There is a table below showing the rumored differences between generations. The rumors about Ryzen 7000 Phoenix don’t stop there. It has also been suggested that PCIe 5 support could also be included, although it has been speculated it will be initially limited with how many lanes can be used. More processing cores may also be on board, with VideoCardz stating that up to 12 cores could feature on Phoenix APUs. As for the iGPU architecture, while Cezanne Ryzen 5000 uses Vega, the above-mentioned roadmap has claimed that AMD will go with RDNA2 (Navi) for Rembrandt Ryzen 6000, and it seems likely that Ryzen 7000 would also feature RDNA2 graphics technology with some additional tweaking. As for a Phoenix release timeframe, we could see it rise from the ashes by the first quarter of 2023, so we still have a whole two years full of rumors, leaks, and confirmations to help build up an accurate picture of what will be Zen 4 mobile. I find it interesting that AMD early rumors contain relatively more real information than Intel's real product announcements actually have in them ---- something about this says that there is more real change going on with AMD than at Intel. Realize that what has already been announced in mainframe has to already exist in the base chiplets and remember those same chiplets are used up and down the entire AMD line up ...... between the different line announcements we are actually getting a good bit of various major features being confirmed directly by AMD. =================================================== Production crunch at TSMC seems to be easing a bit. Some AMD products are in stock again at AMAZON and scalper prices have abated on those items that are "in stock" at AMAZON. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/15/21 at 10:50:39 https://www.tomshardware.com/news/amd-unveils-epyc-milan-7003-cpus-zen-3-comes-to-64-core-server-chips https://www.anandtech.com/show/16529/amd-epyc-milan-review Every mainframe rumor has come true ...... AMD has delivered on what was promised. Now all the other channels (consumer, gamer, laptop) will get these same benefits from the base chiplets being used up and down the AMD lineup. =================================================== Next AMD bump is 5nm (soon) 3nm will be out a year from now but only in selected AMD products. LIE A LOT BIGGER, INTEL --- you have a series of large gaps coming up that you have to make up. http://https://www.hebergementwebs.com/image/af/af9b11275efda92b6514ad821276ebf1.webp/amd-aurait-detrone-intel-in-term-of-desktop-processor-sales-4.we Intel has got some serious looking problems facing them. AMD keeps on outdoing Intel's lies (and yes, Intel lies a lot, actually). Apple M-1 series is flat out kicking on Intel's butt some, too. The new CHROMEBOOKS are actually hurting Intel worse than Apple is, market share wise. Nobody ever puts together market share graphs showing all the players in the different OS systems added up at the same time vs Intel. Likely Intel refuses to cooperate with that sort of exposure as it makes Intel look really really bad. Example: Half of the blue Windows line shown just below is AMD and half is Intel right now --- that is roughly only a 35% of the world market share for Intel. If AMD, Apple and Chromebook each take an additional 10% world market share that puts Intel in something pretty close to the #3 position worldwide. OK, its not likely, but it is within the realm of the possible growth numbers ..... Ring in some other up and coming players like ARM PC chipsets and the new "getting better all the time" RISC-V chipsets, then it only gets worse for the incumbent Intel. http://https://i0.wp.com/www.windowscentral.com/sites/wpcentral.com/files/styles/large/public/field/image/2021/02/os-market-share-2020-idc-geekwire.png?w=1200&ssl=1 RISC-V is a very slow building threat right now, with MIPS now designing and supporting more and better RISC-V processor designs (having just 100% dumped off their own MIPS designs). AMD is doing a better job at getting enough wafers out of TSMC currently so look to see the hurt on Intel increase over the course of this year. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/18/21 at 13:47:05 https://www.cpubenchmark.net/cpu.php?cpu=AMD+Ryzen+9+5900X&id=3870 click on it to see it https://www.cpubenchmark.net/singleThread.html click on it to see it Intel's new Rocket Lake supposedly edges out Ryzen on single core performance (but only on a single core test and only on Intel's favorite test of the moment (a special "single core best of the lot" test that is heavily tweeked for AI usage) and only for a short thermal duration and only on two very specific Intel sku units that were cherry picked for this very purpose). New Intel BS trickery is suspected (see follow on posts for analysis). What is this data good for? Well, it challenges AMD to go beat it, and you know they will, plus an extra 10-15% for good measure. As will Apple's M-2 chipsets coming out later this quarter ....... But, by then Intel will have printed these "new results" on 10's of thousands of product boxes so Joe and Rita Sixpack can continue to read that Intel is still in the lead, still, whenever they go computer shopping. So Intel's whopper marketing lies actually helps motivate AMD and Apple to build "the processor staircase to Heaven" one step height at a time. Also, Intel likes to push out yak yak that they plan to do this and plan to do that, but the generations of stuff Intel promises never gets built as it was never real in the first place. Meanwhile, Joe and Rita never see this stuff ........ click on each item, it is a full page on all the processors that are for sale now https://www.cpubenchmark.net/desktop.html https://www.cpubenchmark.net/laptop.html =================================================== https://liliputing.com/2021/03/lilbits-intels-adaptive-boost-for-rocket-lake-chips-android-11-for-chromebooks.html https://www.anandtech.com/show/16564/intels-new-adaptive-boost-technology-floating-turbo-comes-to-rocket-lake And this, boys and girls, is why AMD provides a very good fan and fin stock cooler system for all their chipsets, and insists that all published testing is done using the stock cooler. AMD only claims what a stock AMD rig can do ...... unlike Intel which lies out their ass all the time using a variety of tricks which include non-standard settings and custom cooling. Lilbits: Intel’s Adaptive Boost for Rocket Lake chips This week Intel introduced its new 11th-gen 14nm desktop processors, code-named Rocket Lake-S. Set to hit the streets at the end of the month, the new processors are manufactured using a 14nm process rather than the 10nm process the company uses for most of its recent laptop chips. But Intel is still promising a modest boost in CPU performance and bigger boosts in integrated graphics. Intel is also promising even better performance for its most powerful Rocket Lake-S chips. The upcoming Intel Core i9-11900K and Intel Core i9-11900KF will support a new feature called Adaptive Turbo Boost, which allows some of the eight CPU cores to run at up to 5.1 GHz simultaneously, assuming they processor is running cool enough to support the feature. AnandTech has a more detailed explanation, including some useful graphics that show how Adaptive Turbo Boost differs from regular Turbo Boost and Thermal Velocity Boost. The tech news and analysis site also already has a review of one of Intel’s less powerful Rocket Lake-S chips, the Core i7-11700K, and it seems like the processor only kind of delivers on Intel’s promises of performance improvements. It all depends on what you ask the chip to do. Intel’s New Adaptive Boost Technology: Floating Turbo Comes to Rocket Lake [AnandTech] Intel’s new Adaptive Boost technology will bring better multi-core performance to the highest-performance Rocket Lake-S desktop processors by enabling selected CPU cores to hit speeds as high as 5.1 GHz simultaneously under the right conditions. Question: how many seconds of this action before the BIOS thermal sensor built into the chipset itself shuts it down for thermal throttling (self preservation) ???? http://https://images.anandtech.com/doci/16564/1-11900KTurbo_575px.png So, it is a another hyperclocking gimmick that only operates in certain circumstances (such as a carefully set up CPU benchmark test) to give out a false impression of real processor performance under those certain specific circumstances. An Intel PC equipped with chilled water or freon cooling might be able to get all 8 cores actually up and running on this trick (aside from the fact that Intel rates the processor off the single one best performing core and the rest are always catch as catch can as far as real performance goes. ::) Using them Intel "Magic Minute" tricks again I see ........ however this time they are only a few magic seconds in duration. USING FAN AND FIN COOLING, THERE ARE CURRENTLY NO REAL WORLD TASKS THAT CAN ACTUALLY USE INTEL'S ADAPTIVE BOOST TECHNIQUES INSIDE THE VERY FEW SECONDS THAT BIOS CONTROLLED THERMAL THROTTLING WOULD PERMIT FOR THEM TO BE EFFECTIVE. Intel Marketing BS, smoke and mirrors, lies you pick the term you would prefer to use. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/22/21 at 13:20:04 https://www.pcgamer.com/intels-rocket-lake-is-looking-red-hotfor-all-the-wrong-reasons/ https://www.anandtech.com/show/16535/intel-core-i7-11700k-review-blasting-off-with-rocket-lake Intel's Rocket Lake is looking red hot... for all the wrong reasons By Jeremy Laird January 29, 2021 Intel’s stopgap 14nm swansong runs hot and hungry. The top Intel Rocket Lake CPU runs at a red hot 98 degrees C and guzzles 250 watts. So says a post on Chiphell, which details the top Core i9 11900K model running a full stress test. The chip was reported running at 4.8GHz all-core, which aligns with the best currently available information on what remains an unannounced, if imminent, new processor. The last of the 14nm line before Intel finally switches its desktop CPUs to 10nm. The latest leak from motherboard maker MSI indicates the Core i9-11900K will run at 3.5GHz base clock, 5.1GHz Turbo, 5.2GHz Boost Max 3.0, and 5.3GHz Thermal Velocity Boost. That’s modern CPUs for you. There’s no such thing as a simple answer to the question of how fast they run. Anyway, if 98 degrees and 250W sounds bad, the chip was reportedly running a full stress test rather than real-world apps. Intel’s existing Comet Lake equivalent, the Core i9 10900K, has been recorded at 93 degrees and 235W running a stress test. Thus, Rocket Lake flirting with the 100 degree barrier wouldn’t be entirely unprecedented. That said, the Core i7-10900K is a 10-core chip, where the upcoming Rocket Lake 11900K will pack just eight processor cores. If it does turn out Rocket Lake runs hot and hungry fingers will be pointed at its origins as a 10nm architecture that has since been backported to 14nm silicon on account of Intel’s failure to get the 10nm up and running reliably. With Intel publicly stating that it still plans to ramp production of its 10nm Alder Lake CPUs in the second half of this year, it may end up not mattering much whether Rocket Lake runs at all, let alone hot. But at best the backported Rocket Lake feels like a stopgap CPU rather than the chip to turn Intel’s fortunes around. ================================================== https://www.anandtech.com/show/16535/intel-core-i7-11700k-review-blasting-off-with-rocket-lake has now published the items that the Intel NonDisclosure permits them to publish. A strong smell of fish is in the air over Rocket Lake as only two items out of the list show to be an improvement at all, the rest are regressions at varying levels. Save your money and keep your old Intel chipset ----- Rocket Lake only offers some mild very specious improvement numbers so far. ================================================== https://www.youtube.com/watch?v=3n0_UcBxnpk "A waste of sand" ???? Nothing advantageous at all ????? No meaningful difference? It is being questioned by this guy that Intel is simply pushing an intentional nothing burger hidden under the very very laggy non-disclosure period to let them get this garbage firmly out into the retail channel before the dirty secret gets out. Intel is trusting that "chipset shortages worldwide" will see these turkeys sold anyway. And Intel is likely right about that ......... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/23/21 at 10:04:44 OK, the big guns won't say a word until Intel releases the embargo on Rocket Lake. Silence remains for another couple of weeks according to embargo rules. The little guns say "Hey, you shipped the stuff out to the public and I bought mine at a retail outlet, so embargo smargo to you, Intel". https://wccftech.com/intel-core-i9-11900k-rocket-lake-desktop-cpu-benchmarks-tested-against-amd-ryzen-7-5700g-ryzen-7-5800x/ Intel Core i9-11900K Rocket Lake Desktop CPU Also Tested Against AMD Ryzen 7 5700G & Ryzen 7 5800X, Unimpressive Gaming Performance It is a long article, basically saying the Intel AI tweeks either work for that test or the CPU simply sux on that particular aspect. Extreme hyperclocking is seen from Intel, and pointed questions are asked about if the standard fan and fin coolers are even being appropriate for Rocket Lake use at all. CPU BIOS Level Throttling is extreme and very very quick to arrive. Some of the standard tests done simply didn't pan out at all vs Intel's marketing claims --- the most egregious being the poor gaming results. Nowhere was a 10-20% advantage found for Intel, ever. Actually, Intel never showed ANY advantage at all, really. Intel's cherry picking of their shipped to the reviewers units really shows here, badly. These are run of the mill as shipped to the public units, and they suck. http://https://cdn.wccftech.com/wp-content/uploads/2021/02/Intel-Core-i9-11900K-Rocket-Lake-8-Core-Desktop-CPU-Gaming-Performance-Benchmarks-vs-AMD-Ryzen-7-5800X-Ryzen-7-5700G-1030x579.png |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/23/21 at 15:13:13 https://www.businesswire.com/news/home/20210323005981/en/Intel-CEO-Pat-Gelsinger-Announces-%E2%80%98IDM-2.0%E2%80%99-Strategy-for-Manufacturing-Innovation-and-Product-Leadership/ Intel outlines 7nm “Meteor Lake” chip timeline, plans for first-party and third-party fabs Intel CEO Pat Gelsinger says it’s on track to tape out its first 7nm “Meteor Lake” chip in the second quarter of 2021, but they won’t actually ship until 2023. SANTA CLARA, Calif.--(BUSINESS WIRE)--Today, Intel CEO Pat Gelsinger outlined the company’s path forward to manufacture, design and deliver leadership products and create long-term value for stakeholders. During the company’s global “Intel Unleashed: Engineering the Future” webcast, Gelsinger shared his vision for “IDM 2.0,” a major evolution of Intel’s integrated device manufacturing (IDM) model. Gelsinger announced significant manufacturing expansion plans, starting with an estimated $20 billion investment to build two new factories (or “fabs”) in Arizona. He also announced Intel’s plans to become a major provider of foundry capacity in the U.S. and Europe to serve customers globally. “We are setting a course for a new era of innovation and product leadership at Intel,” said Gelsinger. “Intel is the only company with the depth and breadth of software, silicon and platforms, packaging, and process with at-scale manufacturing customers can depend on for their next-generation innovations. IDM 2.0 is an elegant strategy that only Intel can deliver – and it’s a winning formula. We will use it to design the best products and manufacture them in the best way possible for every category we compete in.” IDM 2.0 represents the combination of three components that will enable the company to drive sustained technology and product leadership: 1. Intel’s global, internal factory network for at-scale manufacturing is a key competitive advantage that enables product optimization, improved economics and supply resilience. Today, Gelsinger re-affirmed the company’s expectation to continue manufacturing the majority of its products internally. The company’s 7nm development is progressing well, driven by increased use of extreme ultraviolet lithography (EUV) in a rearchitected, simplified process flow. Intel expects to tape in the compute tile for its first 7nm client CPU (code-named “Meteor Lake”) in the second quarter of this year. In addition to process innovation, Intel’s leadership in packaging technology is an important differentiator that enables the combination of multiple IPs or “tiles” to deliver uniquely tailored products that meet diverse customer requirements in a world of pervasive computing. 2. Expanded use of third-party foundry capacity. Intel expects to build on its existing relationships with third-party foundries, which today manufacture a range of Intel technology – from communications and connectivity to graphics and chipsets. Gelsinger said he expects Intel’s engagement with third-party foundries to grow and to include manufacturing for a range of modular tiles on advanced process technologies, including products at the core of Intel’s computing offerings for both client and data center segments beginning in 2023. This will provide the increased flexibility and scale needed to optimize Intel’s roadmaps for cost, performance, schedule and supply, giving the company a unique competitive advantage. 3. Building a world-class foundry business, Intel Foundry Services. Intel announced plans to become a major provider of U.S.- and Europe-based foundry capacity to serve the incredible global demand for semiconductor manufacturing. To deliver this vision, Intel is establishing a new standalone business unit, Intel Foundry Services (IFS), led by semiconductor industry veteran Dr. Randhir Thakur, who will report directly to Gelsinger. IFS will be differentiated from other foundry offerings with a combination of leading-edge process technology and packaging, committed capacity in the U.S. and Europe, and a world-class IP portfolio for customers, including x86 cores as well as ARM and RISC-V ecosystem IPs. Gelsinger noted that Intel’s foundry plans have already received strong enthusiasm and statements of support from across the industry. To accelerate Intel’s IDM 2.0 strategy, Gelsinger announced a significant expansion of Intel’s manufacturing capacity, beginning with plans for two new fabs in Arizona, located at the company’s Ocotillo campus. These fabs will support the increasing requirements of Intel’s current products and customers, as well as provide committed capacity for foundry customers. This build-out represents an investment of approximately $20 billion, which is expected to create over 3,000 permanent high-tech, high-wage jobs; over 3,000 construction jobs; and approximately 15,000 local long-term jobs. Today, Arizona Gov. Doug Ducey and U.S. Secretary of Commerce Gina Raimondo participated with Intel executives in the announcement. Gelsinger commented: “We are excited to be partnering with the state of Arizona and the Biden administration on incentives that spur this type of domestic investment.” Intel expects to accelerate capital investments beyond Arizona, and Gelsinger said he plans to announce the next phase of capacity expansions in the U.S., Europe and other global locations within the year. Intel plans to engage the technology ecosystem and industry partners to deliver on its IDM 2.0 vision. To that end, Intel and IBM today announced plans for an important research collaboration focused on creating next-generation logic and packaging technologies. For more than 50 years, the two companies have shared a deep commitment to scientific research, world-class engineering and a focus on bringing advanced semiconductor technologies to market. These foundational technologies will help unleash the potential of data and advanced computation to create immense economic value. OK, as everybody else leaves 7nm going down to 5nm Intel just now announces a 3 year plan to GET TO DIRECT BURN EUV 7nm by the end of 2023. Intel will arrive at the same time the rest of the world is going down to 3nm ........ Gelsinger's hand is out now and now his hand is deeply embedded in Joe Biden's wallet pocket, starting out with lifting a cool 20 billion dollars FOR NOTHING. This is stupid money spent wastefully, as Intel and IBM never had a clue about how to successfully do 7nm, much less 5nm or 3nm. If the plan was to have TSMC or Samsung build them a turn key plant and park some TSMC and Samsung associates in it to help Intel run it, it could be said to make some sense. This is just plain old fashioned stupid meaningless political pork barrel spending ....... Gelsinger is a smart man, he realizes that he needs to skip over 7nm and 5nm and drop on down to 3nm to get back into the swing of things. However, Gelsinger leads a company that is sorely lacking skills in the lower nanometer ranges. Gelsinger needs to go headhunting, aggressively, for new much better talent and he needs to give these smart newer engineers the 20 billion he just lifted out of Biden's wallet. There are many many ways in which Intel can fail in these big grandiose plans. An issue Intel has is that they have NO history of making a lasting partnership with anybody -- Intel instead has a LONG LONG history of screwing over their partners. I am surprised IBM would even have them back since IBM is one of the first major major screw jobs that Intel did to a partner. As far as Intel becoming a Fab ---- Intel has been there and failed at it twice previously. AS A FAB SIPPLIER, Intel never delivered anything both on time and on price, plus if production was tight Intel would always make their own stuff for themselves first. If you think Intel has changed their tune on building stuff for their customers first, you are DELUDED, badly self-deluded. Biden has just spent 20 billion so INTEL can modernize and increase their internal production outputs. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/24/21 at 10:03:40 https://www.anandtech.com/show/16535/intel-core-i7-11700k-review-blasting-off-with-rocket-lake More Rocket Lake reviews are posted, and they are not good. http://https://images.anandtech.com/graphs/graph16535/121878.png Performance gains range from mediocre to non-existent, and in the worst cases it's actually slower than Skylake, overall it's still slower vs the 5800X despite consuming significantly more power. Interestingly, it also has higher core-to-core latency and a worse performing L3 cache compared to Skylake. Mind you, these reviewers actually got their chipsets from open & free EU STANDARD RETAIL CHANNELS, which shows just how badly Intel had worked to cherry picked the better working samples that Intel had sent out to the main review houses for their reviews. Now you understand why Lisa Su had told the boys and girls last time to just go buy an AMD chipset of the last 5000 series grouping at random out of the retail stream. Lisa Su did not want AMD to play in this presorted Intel BS samples sandbox at all. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/24/21 at 20:50:07 Re-listening to Gelsinger's presentation for a third time, I am struck by just how closely he plans to copy Lisa Su and AMD's chiplet approach to chipmaking. This is the iconic Intel technology stealing method which can be stated as "copy & collaborate, assimilate, modify and then patent the most current Intel in-house modifications by which action Intel assumes functional control of the combined technology that they did not invent". Intel did this to IBM at the start of the PC era, they did it later to APPLE with the Steve Jobs visits, they have done it to Micron over 3DXpoint and now again to AMD Chiplet tech (most recently and most egregiously) as well. Godzilla lives again. Gelsinger remembers very well how Godzilla works and he is using these old Godzilla methods as his business model. Following a classic cut and build strategy, Bob Swan had done all the cutting (all the way down to ground zero) and Gelsinger is now using Biden's incentive money to build a brand new Intel from Bob Swan's leftover ashes up. Gelsinger has a good vision laid out clearly in his head and he is taking steps to make it all come together. Scariest yet, Gelsinger is a good enough engineer and an enthusiastic enough salesman to make it all work ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/30/21 at 21:24:04 https://liliputing.com/2021/03/lilbits-rocket-lake-s-piunara-raspberry-pi-cm4-carrier-board-and-linux-for-old-nvidia-tegra-devices.html You can buy a desktop (or even a laptop) computer with an Intel Rocket Lake-S processor starting today. But should you? The first round of reviews are in, and it seems like Intel managed to pull off some sorts of "occasional" performance gains in single-core performance despite sticking with 14nm. But depending on what you’re asking your computer to do, you might actually get better performance from a previous-gen Intel processor. And of course, there’s always AMD, whose Ryzen chips have been killing it over the past few years. AMD continues to outperform 14nm Intel processors and the existing 10nm Intel processors across the board. http://https://liliputing.com/wp-content/uploads/2021/03/rocket-lake-700x394.jpg When AMD goes 5nm look to see even greater AMD gains over all of Intel's offerings. All the chatter about Intel suing AMD got chopped off cold, suddenly, so something has happened. Intel might have gotten asked by the FCC did they really not remember the terms of the last judgement the FCC rendered against Intel, giving VIA and AMD rights to 32 bit x86 forever? AND, INTEL IS NOW ADVERTISING THAT THEY WILL SELL A 64 bit x86 LICENSE NOW TO ANYBODY WHO WANTS TO BUY ONE, so what exactly the heck was Intel suing AMD over again? "Acting in bad faith" can mean falsely suing over A while trying to get B & C in the settlement, this is one current interpretation of "bad faith". So now Intel wants to peddle AMD's 64 bit x86 tech like they owned it? And this was after saying they were going to cut AMD off from 32 bit 286 generation Intel x86 tech just last week? Intel's lawyers got some stupid directions to go after some stupid things from Bob Swan ....... very very self destructive things no less. ================================================== Intel's Embargo is over, last BIOS updates have dropped from all the major board vendors. the Rocket Lake roll ups are coming in, and it does not look good for Rocket Lake in the main gaming section. http://https://images.anandtech.com/graphs/graph16549/122522.png So, how does Intel move forward? They have lowered the cost of their products as far as they can and are making massive amounts of them, so Intel now has superior pricing and an ample supply of chipsets right now. AMD has the higher prices and somewhat shaky supply lines through TSMC, but AMD is much preferred by hobby PC builders right now. The big "pre-built box peddling" stores seem to belong to Intel right now, as does Dell, Lenovo and HP. Intel will shower you with slightly lowered prices and lots of "admittedly it isn't really so great and it runs really really hot" products for the rest of this year. Intel is making progress, just at a third the rate AMD and Apple and the ARM boys are doing. Speaking of ARM, the next generation of ARM will be BIG, LITTLE and LARGE COMPUTER cores, three separate levels of ARM cores each aimed at their separate niches. Intel will get dozens of extra competitors for their market share ....... When the testing houses roll their overall processor performance rankings and pricing rankings I will post them. Next up in about 7 months we will have AMD 5nm product introductions. We already know 5nm chiplets have been shipped and AMD is working on assembling their first trial finished products for evaluation, so we should be seeing some performance leaks soon. ;D |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/31/21 at 12:53:29 The internet reviewers are being MERCILESS, Intel Gen 11 is pure Intel puppy poo --- scrape it off your shoe soles and then go wash your shoes real good. https://www.youtube.com/results?search_query=intel+gen+11+a+waste+of+sand Jest pull it up and LOOK at the LONG list of pan reviews from all over the web. :o ....... and yes you really can boil a cup of tea off the excessive processor heat Intel produces while not running near as good as AMD does ....... =================================================== INTEL REACTS to the bad press ....... by ignoring Rocket Lake completely now and starting up heavy on the rumor cycle on it's next "great thing" which is due in the first half of next year. Intel has birthed a turd, notice how Intel actively hid the late pregnancy and the deformed birth behind a LONG LONG LONG embargo period. Now the deformed baby turd is going for a secret adoption in Switzerland and will never be spoken of again. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 04/01/21 at 15:09:25 https://liliputing.com/2021/04/amd-ryzen-5000g-cezanne-desktop-processor-details-leaked.html http://https://liliputing.com/wp-content/uploads/2021/04/ryzen-700x394.jpg As spotted by @momomo_us, there are multiple pages on HP’s international websites mentioning the new Ryzen 5000G Cezanne chips. I suspect some of those links to disappear, but for now it seems pretty clear that AMD’s new desktop processor lineup will include these three models: Model Cores/Threads Base Freq Max Freq L3 cache TDP Ryzen 7 5700G 8/16 3.8 GHz 4.6 GHz 16MB 65W Ryzen 5 5600G 6/12 3.9 GHz 4.4 GHz 16MB 65W Ryzen 3 5300G 4/8 4 GHz 4.2 GHz 8MB 65W These new chips all feature the same Radeon Vega graphics as their Ryzen 4000G predecessors, but the new models have twice as much L3 cache and higher base and boost CPU frequencies. More importantly, thanks to their new Zen 3 CPU cores, they should bring a boost in IPC (instructions per cycle), which means that even running at the same speed, they should be about 19-percent faster than previous-gen chips. Able to consume up to 65 watts, they should also be able to outperform their laptop counterparts like the 15-watt Ryzen 5000U and 45-watt Ryzen 5000H series processors, while still offering a more energy-efficient design than AMD’s Ryzen 5000 “Vermeer” desktop chips which also use Zen 3 CPU cores, but which have TDPs up to 105 watts and which do not include integrated graphics. I seem to detect a pattern where Intel seems wants to compare itself to THE WRONG AMD PRODUCT CLASS in order to create a cheap "victory for Intel". This is explained partially by the fact that Intel draws over twice the watts of power to do a given job compared to "an equivalent" AMD product. And the core counts don't match up either. We have seen Intel reach back in time to use data from obsoleted AMD products trying to do a match up head to head that actually did happen to have a core count match up. Plus, AMD is always changing and getting better at the chiplet level ---- so when Intel does try to develop a comparison set up to prove they are better then the AMD part referred to likely gets improved again before the running late Intel competitor can actually make it to the starting gate. ;D AND THIS IS NOT EVEN INTO THE NEW 5nm BASED NEW GEN AMD PRODUCTS YET .......... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 04/02/21 at 10:54:26 https://liliputing.com/2021/04/report-googles-whitechapel-processor-will-power-the-pixel-6-smartphone.html Google and Samsung start building custom Google ARM phone processors on Samsung lithography equipment. http://https://liliputing.com/wp-content/uploads/2021/04/pixel-6-418x500.jpg We first saw reports that Google was developing its own custom processors about a year ago, but there hasn’t been much news since then. Now 9to5Google says it’s reviewed “documentation” that confirms the Pixel 6 phone will have a Whitechapel processor when it ships this fall. Other details include references with something code-named “Slider” which may be another name for the system-on-a-chip and which also apparently confirms that Google is working with Samsung to develop its new processor, since other Slider-related projects include references to Samsung Exynos processors. There are references to two phones, code-named “Raven” and “Oriole,” one of which will most likely be called the Pixel 6, while the other may be a more affordable model like last year’s Pixel 4a 5G. It’s unclear at this point what, if any, advantages the new GS101 processor will offer over Qualcomm’s latest chips. But by developing its own processors in-house (even with assistance from Samsung), Google will have more control over the features of the chips used in its smartphones and other products. And that could help Google ensure that hardware and software work together efficiently to offer improved performance, much the way Apple does with its chips for iPhones, iPads, and more recently Macs. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 04/02/21 at 17:37:20 https://www.windowscentral.com/intel-12th-gen-chip-details-leak-online http://https://www.windowscentral.com/sites/wpcentral.com/files/styles/large_wm_brw/public/field/image/2020/09/intel-core-i5-10600k-2.jpg What you need to know Leaked slides detail Intel's 12th Gen upcoming Alder Lake-S CPUs. The CPUs offer 20% better single-thread performance and double multi-thread performance over their predecessors ........ is this actually the same 20% we were promised but we didn't get with Rocket Lake? The Alder Lake-S processors should launch in the second half of this year. Details about Intel's upcoming Alder Lake-S processors have leaked online. A report by VideoCardz shares slides detailing Intel's 12th Gen CPUs. According to the report, Intel claims 20% single-thread performance improvement on its Golden Cove core design. The Alder Lake processors will also allegedly be twice as fast in multi-threaded workloads. The slide mentions Intel's new Gracemont cores alongside the improved multi-thread improvements. VideoCardz points out that it's not known if these improvements are in comparison to Intel's Rocket Lake or Tiger Lake processors ........ is this actually the same 20% we were promised but we didn't get with Rocket Lake? The Alder Lake CPUs feature Intel Hybrid Technology with small high-efficiency cores and big high-performance cores. The Alder Lake CPUs will support PCIe Gen 5 and Gen 4 as well as DDR5 and DDR4. On the mobile CPU side of things, the Alder Lake processors support LPDDR4 and LPDDR5. A separate report from VideoCardz reveals that the Alder Lake chips will support LPDDR5X. VideoCardz also shared details about the Intel 600 series chipset. The desktop CPUs will support dual-channel DDR5-4800 memory. Motherboards that are DDR-4 capable will support up to 3200 MHz modules, while high-end Z690 motherboards will offer DDR5 support, according to the report. The CPU will support PCI Gen 5 and have 16 PCI Gen5 capable lanes. It will also have 4 PCI Gen4 lanes. This is a set of 10nm big little cores "done Intel style" presentation. The very first set of big little that were sent out for samples did not impress anybody and the Rocket Lake back-port was pulled forward to distract us all from Intel's past 10nm mistakes. The back-ported 10 to 14nm Rocket Lake design really was a "waste of sand" beyond just acting as a distraction. So the 14nm distraction flopped strongly, so here is the 10nm main event served up again only delayed a half a year so it could finish cooking some more up on the range top. ;) Intel has actually picked up on their pace a bit, two-three abysmal failures inside one calendar year is twice the "accomplishment rate" Intel has done previously. ;) Intel seems to be committing to 4 threads per core future now, so Microsoft needs to finish writing it into their OS system code for Intel to be able to use it. AMD has already done this trick for AMD's 5nm Zen 4 generation, so it is time for MS's Windows OS to catch up to the new reality of 4 threads per core. Intel is now talking about using new nomenclature for their chipset feature sizes. Intel wants to call their stuff 5nm and 3nm just like TSMC does, whether it is or isn't, or if it acts like it is or not, either one. Whatever little meaning you could take from Intel's product names, part numbers and descriptions gets even weaker still, accordingly. AMD, however, is really in reality going down to TSMC 5nm and Intel simply cannot stand getting left behind at 10nm yet again ...... Take a close look at all the components tacked on the back of the chipset in between the pins -- Intel big-little apparently takes two full sets of these components (one for big, the other for little) to make it work right. :P Looks like Intel Gen 12 will require a new motherboard and a bigger processor socket. :P |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 04/04/21 at 23:16:12 Now for the taking of the deep breath and doing the "considered retrospection" of things in general. I find I don't like playing computer games nearly as much as I used to. My reflexes are slower now and my finger accuracy on whatever controller system I use has declined greatly. My right hand little finger likes to jam in the curled position now and the knuckle and first finger joint on my right hand little finger requires manual intervention from my left hand to get it un-clenched and popped back into working correctly. I would jam up several times on a level and get tend to get killed because of it, so that did get kinda old pretty quick. Since I am not a PC gamer any longer, I find my 15 year old used Dell office box still works just fine for all I do with it. Running Linux Mint Mate is still quick and smooth, so that is good too. Can't shoot, can't game ........ but I can still grow me some tomatoes though. =================================================== Following tech world is still fun, especially since Intel has a competent leader again and Intel is gonna try to actually COMPETE again. And yes, I do realize that Intel will still carefully complete every program that was started by Gelsinger's predecessors (some of which originated back 7 years or more ago). Intel is Intel, and every thing I have called a "distraction" or a "placeholder" recently was intended to be groundbreaking back when it was actually begun. "Overcome by events" is a phrase that does apply here, I think. The times caught up with the Intel plans and rendered them passe' before they were even executed for the first time. We are on try #2 and try #3 on some of these Intel plans, with some very mixed final "improvement levels" being seen. ==================================================== Intel's proposed 10nm Mainframe and home workstation chipsets have now had samples sent out and the benchmarks are popping up all over the place. Results are disappointing to say the least. 10nm Intel (whichever name or flavor you want to talk about) does not compete well at all against the current Ryzen 5000 generation which has also been RELEASED and has been fully benchmarked. Intel is now selling and can supply a volume of chipsets that are "close to competitive" to the previous Ryzen generations. In some use cases, if you can't buy a current AMD you are actually better off performance-wise to keep your old Intel chipset for a few more years, then buy the current best AMD you can afford. Rule of thumb, if the new processor has fewer cores than what you have now, you should stick with what you've got now. COUNT THE CORES OF WHAT IS ON THE TABLE BEFORE PLUNKING DOWN YOUR MONEY ........ In all cases, MAKE THE EFFORT TO GET THE BIOS UPDATE FOR YOUR CURRENT MOTHERBOARD as the AI improvements that are built into that BIOS update are worth more in % performance improvement than that new 10nm Intel chipset will actually give you. THE DEMAND FOR NEW UNITS IS SO HIGH ALL OF AMD'S OUTPUT FOR THIS YEAR IS ALREADY ORDERED AND IS ALREADY SOLD. All the recent AMD wafer increases from TSMC were soaked up instantly. Very quickly the same thing will be true of the #2 choice, Intel. Demand levels are insane and are not likely to decrease for the rest of this year. Chromebooks as a "fill in the slack" option will see another very good year. Intel has dropped their prices some and Intel is working hard to move out as much old "previous generation" warehouse inventory as they possibly can move in this unusual seller's market. Once again, if you can wait, do so. If your machine is old, and you do not game much you really don't need a new machine. If you use a spinning platter hard drive, your very best bang for the buck is subbing in a modern cheap SSD drive as your boot drive. 5nm AMD Genoa and the Intel response to that that set of processors are coming, and will raise the bar ~30-40%~ in absolute performance (all AI boosts are included in this general figure). |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 04/10/21 at 04:58:56 Why wait until next year to buy a new PC? AMD is stepping down to 5nm and Intel is stepping down to 10nm and starting to use TSMC 6nm (built at TSMC, not Intel). This is assuming TSMC wishes to support a competitor fab company called Intel Foundry Services ..... a brand new competitor who has just announced their plans to take TSMC's business away from them. Don't think TSMC isn't smart enough to listen to AMD and to sell that 6nm wafer volume to AMD instead of to Intel ....... Throughput improvements next year will be 30-40% roughly speaking, with PCIe at gen 4 and gen 5 (depending on when you buy it). 4 threads per core SMT will be available. 8-16 data channels will be available. Memory speeds and throughput will double. AI will be more capable and more accepted in OS and programs in general. Core counts from Epyc mainframe chipsets will roll down into upper level workstation chipsets, upper level workstation will edge down into homeowner PC ranges. All of this will be driven by the MUCH SMALLER and much faster and more efficient 5nm chiplets AMD and Xilinx will have completed their merger by next year and several new conjoined technologies will become available for the first time in consumer PCs. More competition will be available to drive more progress in an even faster manner, with ARM PC chipsets, Apple chipsets and new Intel made at TSMC chipsets ...... What you buy next year will be that much better than what you can buy today. Cheaper, too. Nothing available right now from Intel tests to be much better than the Intel you already own -- hunt up the BIOS updates for your motherboard and install them to get over 50% of the real world benefits of a new Intel machine (you get all the AI tricks in other words). Note please how drastic the improvements have been since Ryzen came upon the scene. Intel had kept everybody parked at a 4 core product for over 10 years at that point in time. http://https://cdn.mos.cms.futurecdn.net/oHsVVXYo8rKt3ZstE6dXvZ-970-80.png.we More on the ongoing struggle to find more AMD chiplet wafers ....... As you view the chart above, realize that AMD has already sold all of what they can make with their scheduled 7nm TSMC chiplet wafer supply. As such, AMD is seeking MORE WAFERS OF CHIPLETS and has told TSMC to either provide these additional wafers of chiplets or the contractual second source clauses will have to come into effect. It is rumored that TSMC has already bumped AMD's 5nm allocation as high as they can and has told AMD that if AMD wants any more wafers they will have to come from the new 6nm process lines that the old TSMC 7nm lines are being converted to as we speak. Now we will see the TRUE FLEXIBILITY offered by the AMD chiplet production system. AMD already has 6nm completed processor designs on tap and once the actual 6nm chiplets get made, tested and binned AMD can then select which products will be best able to use them. All of the 7nm layouts can take a 6nm chiplet with a little room to spare (folks argue that 7nm++ is actually a near to 6nm to begin with) and AMD can certainly maximize this short notice chiplet change very easily with their existing processor designs. Next, AMD has a flow of new products coming on that can be adapted specifically to use these new 6nm chiplets once they actually can get them in larger volumes. Lastly, Intel has already announced they are going to TOTALLY SHRED the whole 10-7-6-5-3 nano-meter nomenclature thing inside the next 6 months, calling their 10nm product the Intel 6nm or 5nm or 3nm depending on just how much they want to lie about it. AMD's minor 7nm++ to 6nm cheat is indeed a very minor sin compared to Intel's contemplated goings on. AMD simply needs much greater wafer output numbers to continue taking more of the 50%+ market share graph zone away from Intel. TSMC wants to keep that AMD business intact and they had long promised AMD to be able to make enough wafers of AMD chiplets to do just that. ::) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 04/11/21 at 04:13:53 https://hothardware.com/news/amd-roadmap-ryzen-6000-zen-3-apus-rdna-2-gpu-ddr5 Leaked AMD Roadmap Forecasts Ryzen 6000 Zen 3+ APUs With RDNA 2 GPU And DDR5 Support Confirmation of the rumor that AMD will go for Zen3+ generation using TSMC 6nm processes while still working with 5nm and 7nm units "as available" remembering always that AMD always needs MORE AND MORE AND MORE WAFERS OF CHIPLETS, and AMD needs them very badly so AMD can continue taking more and more market share away from Intel. A rumor that refuses to go away is that AMD will be refreshing its current-generation Zen 3 processor lineup with a new Zen 3+ stack. Apparently this will happen across the board, and not just on the desktop. Serving as further evidence of such is a newly leaked roadmap that points to an upcoming lineup of laptop APUs based on Zen 3+, with support for new features. What you are looking at below are laptop APUs broken down into 4 different segments, including high power (HP, 45W), low power (LP, 15W), ultra low power (ULP, 9W), and extreme low power (ULP, 4.5W). If the roadmap is accurate, a Ryzen 6000 series will debut next year, with DDR5 and LPDDR5 support in tow, and beefier onboard graphics. For what it's worth, well known leaker @kopite7kimi commented on the roadmap saying, "Someone told me this is true." It is certainly plausible, as well as a little bit exciting—moving on from Vega to Navi 2 for graphics chores could result in some interesting and powerful designs, both in the laptop space and in the mini PC department. Right click on the graphic and view it in a new tab to see it all at once, or else slide it over so the relevant Rembrandt-U areas are visible. As TSMC rolls their existing EUV direct burn 7nm++ lines over into EUV direct burn 6nm this rolling refurbishment is forcing a natural evolution of the Ryzen Gen 5 product line. http://https://hothardware.com/Image/Resize/?width=1170&height=1170&imageFile=/contentimages/NewsItem/54772/content/big_amd_cpu_roadmap.jpg PC SHIPMENTS ARE UP 55% ALREADY YEAR ON YEAR, and last year was a strong one to use as your base line. The company (and their supply chain) that can ship more product right now wins, in other words. http://https://liliputing.com/wp-content/uploads/2021/04/canalys-q1-pc-shipments-700x394.jpg Intel has proven to everybody it does not have to be particularly great stuff that Intel is shipping, Intel just has to have it available to ship it right now. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 04/12/21 at 15:44:38 https://nvidianews.nvidia.com/news/nvidia-announces-cpu-for-giant-ai-and-high-performance-computing-workloads MEET GRACE, INVIDIA'S COMBINED TECH ARM SUPERCHIP DAUGHERBOARD ARRANGEMENT 84 ARM Compute units pared with an Nvidia superchip for AI, learning and graphics. http://cms.ipressroom.com.s3.amazonaws.com/219/files/202104/b3cfff26-6195-4f21-a7cc-55f7f16af7a6 This is the sort of stuff you would see NVIDIA making if they were allowed to actually take control of ARM Holdings. Lots of Crushed Competition lying in the gutter, in other words, with NVIDIA wearing the One Ring and ruling over all of the world. GEAR UP, AMD THIS IS WHO YOU ARE COMPETING AGAINST NOW YOU NO LONGER HAVE A COMFORTABLE LEAD AGAINST "BARELY BEGINNING TO MOVE AND NOT REALLY TRYING VERY HARD" INTEL, YOU ARE BEHIND THE 8 BALL AGAIN AGAINST A VERY FAST MOVING AND HARD HITTING ARM-NVIDIA. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 04/16/21 at 10:21:45 https://liliputing.com/2021/04/chinese-chip-maker-loongson-has-designed-its-own-cpu-instruction-set-architecture-isa.html Chinese chip maker Loongson has designed its own CPU instruction set architecture (ISA) http://https://liliputing.com/wp-content/uploads/2021/04/3a5000-492x500.jpg Chinese chip maker Loongson used to make processors based on MIPS architecture, but MIPS isn’t what it used to be and the company that developed the technology has even pivoted to making chips using the RISC-V instruction set. Now Loongson is trying something new: rather than adopt RISC-V, ARM, or x86 architecture, the company has developed its own ISA (instruction set architecture) called LoongArch. The move makes a certain kind of sense in the current political climate. With the United States in the habit of hitting Chinese companies like Huawei and Xiaomi with trade restrictions, it could be risky for Loongson to adopt architecture that relies on contributions from Western companies. Building its own ISA gives Loongson more control over the technology used by its future chips which could be better for the company’s long-term prospects. is no simple undertaking But designing a new ISA from scratch and it’s unclear how competitive the first-gen LoongArch chips will be. Loongson says it’s already taped out the first of its next-gen processors, the Loongson 3A5000, which should be available later this year so we may not have long to wait to find out how they compare to previous-gen Loongson chips based on MIPS architecture as well as competitors based on other ISAs such as x86, ARM, or RISC-V. I betcha they copied chunks of MIPS and chunks of Linux and chunks of DSB and cobbled it all up together. Mostly MIPS though as China has some residual legal rights to that OS currently as they owned it several years ago and wrote themselves a perpetual license that still remains. Intel and AMD and MS will be slightly injured by the lost market share when the Chinese Gov. requires all Chinese business to use this home grown OS kludge up job. This has happened before and it only lasted a year or so before repeated ongoing compatibility issues killed the national sanctioned Chinese clone OS system. Back and forth license and patent violation suits will follow later on going both ways, of course. If anyone cares enough to do so that is ........ ::) Failure to sue means you are giving away your IP to Communist China, which is not a good idea because your will see it in their hands --- forever. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 04/19/21 at 20:38:31 https://www.gsmarena.com/mediatek_first_to_relase_a_4nm_chipset_already_has_orders_from_oems-news-48730.php http://https://fdn.gsmarena.com/imgroot/news/21/01/mediatek-dimensity-1200-realme/-1200/gsmarena_000.jpg According to two separate sources, MediaTek is going to be the first chip maker to release a 4nm chipset. Production is supposed to start either at the end of this year (Q4 2021) or at the beginning of 2022. In fact, some major OEMs have already taken orders for MediaTek's high-end SoC. We are talking Oppo, Xiaomi, Samsung, and Vivo. The downside is that MediaTek has secured the 4nm order from TSMC at a big cost - around $80 per unit compared to the current high-end 5nm chipsets that cost between $30 and $35. A famous Chinese tipster claims that MediaTek's hardware will be able to challenge today's Snapdragon 800-series and also reduce power consumption considerably. It's all just rumors for now, of course and we will have to wait almost a year to see if MediaTek beats everyone to the finish line. As TSMC gets a node working well, they shrink it a nanometer or so and sell the production space all over again for a more premium price to somebody else. Intel hates this, because they have carefully build up their current set of Intel lies and BS PR based on the status quo as Intel knew it back when they made their stuff up, then TSMC and Mediatek have the gall to go and shrink right out the bottom of that status quo leaving Intel just clutching a handful of expired vapor. Aggravating, to be sure. 8-) =================================================== TSMC has posted a break-down of what/where they are spending their 100 billion in Cap Ex new fab money upon. The vast majority of the spend is in new 3-2nm capable dual focus direct burn lithography rigs most of which get parked in Taiwan. TSMC is buying their future, in other words. Some of few of these rigs are for fabs here in America, some lesser few are intended for Europe, but most are to be based in Taiwan. =================================================== https://liliputing.com/2021/04/lenovo-launches-three-new-ryzen-5000-laptops-in-china.html Lenovo is launching new 14 inch, 15.6 inch and 16 inch laptops for the Chinese market under the Lenovo Xiaoxin Air and Xiaoxin Pro brands. Most of the new laptops are powered by AMD Ryzen 5000U or Ryzen 5000H processors, and while it’s unclear if any of these specific notebooks will be sold outside of China, it seems like a safe bet that Lenovo will offer similar devices internationally at some point. Similar to the Ryzen 4000 generation, Ryzen 5000 is rolling out in the same form of waves. Mainframe Ryzen 5000 Epic is out now and has been reviewed ---- it kills all Intel Xeon products by at least 25%. (35% is seen in some cases where the incumbent was old and slow) Laptop Ryzen 5000 U and H series processor are out now at similar performance gaps above Intel Tiger Lake, but once again AMD laptop processors are only supplied at the OEM level (built directly into laptop products). Some laptops have better cooling systems built into them and will perform better than the ones with mediocre cooling systems. When they arrive later on this summer the Desktop Ryzen to Intel Rocket Lake performance gap will be even larger since Intel Rocket Lake functionally lost two cores in their efforts to compete by running fewer cores at insanely way high power draw levels (and at insanely way high produced heat levels using a reduced core count so as to fit inside the existing socket size parameters) due to the old 14nm lithography being that much larger, power hungry and that much less heat efficient. Major Intel stock holders are coming across as quite upset by this clumsy retrograde Rocket Lake "fix" by Intel. Intel is getting told bluntly by their controlling stock holder blocks TO GET OFF 14NM IMMEDIATELY AND TO GET THE CORE COUNTS AND CURRENT DRAWS COMPETITIVE with AMD and to do it ASAP. Issue remains that Intel 10nm simply isn't ready at all for desktop production numbers and that Intel has just this week committed to the Biden Administration they will run tons of various automotive chipsets inside 6 months to bail out the automotive industry. Big conflict, there ....... :o :o ::) ::) Intel is now caught between two masters, and most likely Intel will not satisfy either of these masters completely. Splitting up Intel's attention and efforts right now puts "the Intel Recovery as planned" at risk as Intel now has a larger chance of failing that is more than twice as large as before. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 04/24/21 at 05:35:05 https://www.extremetech.com/computing/322120-apples-m1-positioning-mocks-every-x86-cpu-amd-and-intel-have-ever-launched There Are Still Higher-End CPUs Coming from Apple The rumors of higher-end M1 configurations aren’t wrong just because Apple is using the same CPUs for this set of iMac refreshes that it launched in 2020. Unless Apple intends to drop the Mac Pro altogether, it will have to build a higher-end chip to support professional users. http://https://www.extremetech.com/wp-content/uploads/2020/11/Apple_m1-chip-cpu-power-chart_11102020_big.jpg.large_2x-640x360.jpg But outside of that, Apple is selling a single CPU across a wider range of products than any competing Intel or AMD CPU has ever sold. This speaks volumes as to what Apple believes it has its hands on, namely: a CPU fast enough at the quad-core level — because, scaling-wise, the M1 is effectively a quad-core chip, with four low-power cores to handle low-power workloads and provide a little extra performance boost — to address a huge range of markets, while drawing so little power, it can also be sold in a laptop. Apple has birthed the very first "do anything" chipset --- and we will be looking hard to see what they will uncork at 3nm (12 cores) next summer. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 04/25/21 at 04:29:03 DEALING WITH UNCERTAIN SUPPLY LINES AND TESTING LIES https://www.pcworld.com/article/3613156/intel-rocket-lake-s-vs-amd-ryzen-5000-which-should-you-buy.html This PC World analysis covers the released AMD chipsets beating the released Intel Tiger Lake and Rocket Lake chipsets for compute performance (+ 20%) and for energy use (by a factor of TWO, my goodness). Then comes the glass of shockingly cold ice water that PC World then dumped over the whole thing. You can only buy what you can find to buy this week and that means you don't get to do the match ups the magazine guys looked at, you only get to buy what you can find to buy. Intel has just agreed to divert some sizeable amount of their output to support the American Automotive Industry, and Intel is RIGHT NOW starting up their new Fab business which will remove even more PC capacity and cause larger PC shortages beyond what Intel has been showing so far. AMD's TSMC is being affected by water shortages and ongoing power outages in Taiwan, and AMD will see supply shortfalls accordingly. TSMC is building their own water plant from scratch that will make pure water from sea water or from brown waste water ---- yep, the seasonal water shortage thing is that bad. The new water plants and the new TSMC foundry buildings are two years out, thereabouts ...... but Monsoon rains will come before then. Spending your time comparing test results right now is kinda Moot, in other words. AMD is once again walking away from Intel in power consumption and in general processing throughput capacity and in pure processor efficiency per clock cycle. Biden has put ankle shackles on Intel's progress and new capacity plans, so AMD is fixed for Intel whupping for a while now thanks to the President. Apple and ARM and NVIDIA now constitute AMD's current main competitors. Intel focuses only on Gaming, and very clearly shows that Intel is willing to spend ridiculous (over 2x) amounts of wall socket power to eek out some very minor claimed gaming wins. Extreme Processor Heat Output at these levels means all of Intel's hard use testing is done under cooled water jacket cooling systems, and this is not yet being reflected by any real world production unit cooling systems so this gets a big "What the flock, Intel?" response from simple minded folks like me. AMD, running happily at less than half the power that Intel currently requires to do a given task is still using their stock AMD supplied fin and fan cooling set ups very successfully for all testing. If the AMD units were run under Intel's required water jacket cooling systems, then the super cooled AMD units would simply crush Intel's very best products on all the tests on all the fronts and be doing it continuously all of the time, too. Sounds like AMD can thump Intel at will and at any time, pretty much. Just change out your AMD stock fin and fan cooler over to an Intel chilled water type system and change your AMD BIO settings to water cooling. So, testing methods between the companies are very unequal, as are the production cooling requirements needed to make those Intel testing outputs come about. AMD is honest at least, and supplies a stock AMD fin and fan cooler that is used in all their AMD testing. Intel is lying big time with their cooling systems, and by using crooked little marketing companies as Intel's front men for making up these big testing lies ......... and Intel is lying so much a very big big way right now that I will simply say all published Intel this and that test results and performance information is pretty much worthless right now. Until Intel supplies an "in the box stock cooling system" that can match their bogus testing results on a real world end user machine, Intel is simply puffing some little bitty smoke rings right up your nose. http://https://t4.ftcdn.net/jpg/01/38/32/89/240_F_138328988_OO6VuPyQgANjjNnjWHGXC7liXcF5Bf9P.jpg |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 04/28/21 at 19:21:03 Time to cover the "everything" listing since Intel has gotten their OVERCLOCKED, HYPER VOLTAGE 2x DOUBLE CURRENT DRAW chipset ranked in the "everything" listing. Wowsers, Batman ....... can you actually buy a machine built with that 5.2 gighz OVERCLOCKED Intel thing? https://www.amazon.com/Intel-BXRTS2011LC-RTS2011LC-Liquid-Cooling-Kit/dp/B006588Z7G/ref=cm_cr_arp_d_product_top?ie=UTF8 http://https://images-na.ssl-images-amazon.com/images/I/31qzB4vUmYL._AC_.jpg One with the special high output power supply and the super special Intel chilled water cooling system to handle all of the heat the big Intel processors produces? If you can't get the out of stock special chilled water cooler right now, does it even count as "real" as a product, or is it just some more BS PR from Intel? http://https://cdn.mos.cms.futurecdn.net/oHsVVXYo8rKt3ZstE6dXvZ-970-80.png.we |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 05/03/21 at 19:03:38 I don't think Intel's new CEO Gelsinger is a very happy camper right now --- he thought he had got into Biden's wallet for $20 Billion dollars to build him a brand new world for himself and he was working on the EC for an additional $10 Billion for their matching contribution ---- when TSMC's CEO (the ingrate) simply announced they were less than six months away from fulfilling ALL automotive back orders and that TSMC was forming up an Automotive Only plant with dedicated lines and a set of design engineering centers located in America and Brussels to directly support that business. He also said that DELAYED ORDERING was the root cause of the Automotive Plant shut downs and with the due diligence scheduling of automotive chipset orders there would not be any automotive shutdown issues right now ....... He then offered a plan to get everybody current inside 4 month period if they cooperated with scheduling their needs with TSMC. Hey, spend 30 billion dollars and 4 years to fix the problem, or work with TSMC on proper scheduling and have it fixed inside four months ...... hmmmmm ..... lemme think on that a bit. Based upon this, Intel will likely not get the huge amount of funding they were seeking, but will instead get some very much smaller amounts to rework some existing older Intel lines and to also rework some process lines at various other older chipmakers as all the Automotive needs are all older (obsoleted) lithography processes such as what IBM's legacy Global Foundries specializes in. Please remember, IBM still makes their own chipsets for their own uses (read the next post). |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 05/06/21 at 09:19:58 https://liliputing.com/2021/05/ibm-reveals-breakthrough-2nm-chip.html http://https://liliputing.com/wp-content/uploads/2021/05/ibm-2nm-1-625x500.jpg https://youtu.be/HD5KbeR5mtc it is a video, so click on it IBM has a working 2nm chipset built on its own state of the art ASML dual beam direct burn equipment of which they own 2 at this time. This lithography node is 14% smaller than the upcoming chips based on TSMC’s 3nm process, which are due to start shipping some time next year. IBM claims that the performance and efficiency gains over 7nm are substantial. When computational muscle is the top priority, its 2nm performs 45% better. When extended battery life is more important, IBM’s 2nm is capable of delivering performance on par with a 7nm chip while using about 75% less power. This means potentially double the battery life on a phone. The wafer shown has a big Power PC chipset on it, a reminder that IBM still sells Power PC products for mainframe uses. IBM and Samsung both did their work using new gate all around techniques instead of modified FinFET like TSMC and Intel uses. IBM is using carbon nano sheets for their surround gate material. Very energy efficient and very modern tech coming from IBM, here ..... Kudos to IBM research. Impressive, really. IBM Research is still a "know how powerhouse" which is why Intel's Gelsinger had asked IBM specifically for their tech help several months ago. IBM Research was also part of Biden's wonder team that he put together last month. If Biden's team stays fully USA and stays cohesive, then they can do some great things together. =================================================== TSMC has announced a quick 50 billion dollars being spent on TSMC privately owned water purification and power plants to make them independent of Taiwanese national power fluctuations and get rid of the nagging Taiwan general monsoon rain pattern dependencies. TSMC is acting to defend their #1 position as the world now sees them as a dangerous "weak link" in their supply chain of their own economies. China has destroyed whatever blind trust the US and EU had put in TSMC by flying flights of Chinese dive bombers right over the main TSMC plant complex, showing the world that TSMC had been functionally under their Chinese Military's thumb all this time in the clearest way possible. With just a little minor intentional Chinese Military sabotage activity and good 'ol TSMC would (at China's whim) be able to functionally shut down both the US and EU economies. :P Both the EU and the USA will take action to prevent being shut down by anybody, TSMC included. ================================================== TSMC has begun talking about 1.5nm chipsets as their next big thing after 3nm going into main production (like three years from now running on the current 2 years between generations). Intel/IBM will do 2nm and TSMC will do 1.5nm and Samsung will do 1nm. ::) You know how it goes ....... ..... and of course Intel's 2nn will claim to be more dense and "smaller" than Samsung's 1nm ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 05/10/21 at 08:06:07 https://liliputing.com/2021/05/huawei-launches-a-linux-laptop-with-an-arm-based-kirin-990-processo.html Huawei launches a Linux laptop with an ARM-based Kirin 990 processor by Brad Linder Posted on May 10, 2021 at 10:50 am Huawei launches a Linux laptop with an ARM-based Kirin 990 processor Over the last few years Chinese electronics company has been reducing its reliance on tech from other countries in response to trade restrictions imposed by the US. That’s meant developing a home-grown Android alternative for smartphones (albeit one that’s largely based on Android so far). Now Huawei has launched its first laptop that doesn’t feature an Intel or AMD chip. The Huawei Qingyun L410 is powered by Huwaei’s own Kirin 990 processor, an ARM-based chip that was initially developed for smartphones and tablets. http://https://liliputing.com/wp-content/uploads/2021/05/410-700x421.jpg Laptops with ARM-based processors have become more common in recent years, now that Chrome OS, Windows, and macOS all support the architecture. But like I said, Huawei is trying to move away from using technologies from companies like Microsoft and Google, so the Qingyun L410 ships with a Linux-based operating system called Unity OS, or UOS. The operating system, which is developed by UnionTech, was created in 2019 as part of the Chinese government’s push to move away from reliance on Windows for computers used in the domestic market. As for the laptop, ITHome notes that aside from the processor and operating system, it looks a lot like the Huawei MateBook 14. Features include a 14 inch display with a 3:2 aspect ratio, a hidden camera that pops up at the push of a button, a fingerprint sensor, and a security chip. The notebook has 8GB of RAM, up to 512GB of storage, and it measures about 0.6 inches thick and weighs about 3.3 pounds. While the Qingyun L410 is unlikely to be sold outside of China anytime soon, it’s the latest example that it’s getting easier for laptop makers to skip the Wintel duopoly that has dominated the PC space for decades. You will see more ARM based machines like this one, as the ARM as main processor ice has been truly broken now and full Linux support is available to keep these machines running for decades and decades ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 05/11/21 at 10:15:57 https://www.google.com/search?q=tiger+lake-h+processors+with+45+w+tdp&oq=tiger+lake+&aqs=chrome.1.69i57j69i59l2j0i67j0i67i131i433j69i60l3.7121j0j7&sourceid=chrome&ie=UTF-8 This contains a world of testing and other stuff about Tiger Lake H processors in here ---- the gist is that Intel has apparently learned more about how to deal with their heat issues and their low battery life issues (or so they claim) and are now saying Intel can beat Apple M-1 and AMD's Ryzen 5000 chipsets in "real world situations". ::) Major Laptop Magazine Reviewers disagree strongly, but these reviewers factual opinions are not printed on the Intel Inside product boxes, now are they? Intel's semi-bogus exaggerated claims from their paid review houses always get printed right there on the outside of their product boxes where Joe and Rita Sixpack will read them while computer shopping at Best Buy and Walmart ....... This particular point in time is approximately two months away from release of brand new generations of processors from Apple and AMD, so this may be simply another of Intel's carefully timed "temporary claims to fame". ::) (lots of boxes of fresh built Intel units need a brand new set of Intel lies to print on them, in other words) :P You have 3 players now stepping over each other to make progress up the staircase now, so the total progress rate should pick up the pace a wee bit. Intel is beginning to slowly claim some serious progress with their 10nm, while AMD and Apple are at 5nm right now with concrete plans to go to 3-2nm next year. REMEMBER, Intel uses twice to 3 times the power that the other need to do the same jobs and Intel does have some fairly serious issues getting rid of all that heat that comes along with that sort of heavy power usage. All of the players can still sell everything they can build right now ---- it is still a seller's market for everybody at the moment. Chiplet wafer allocation shortages are all that is limiting AMD right now, they simply cannot get enough chiplet wafers out of TSMC to assemble all the products they could sell if they did have larger chiplet wafer allocations flowing at TSMC. AMD will take no greater "larger slice out of Intel's old market share" until they can get the more than 10's of thousands of additional wafers full of TSMC chiplets that are needed to do so. On the converse side of the coin, Intel is still able to sell all of their old style stuff out of their warehouse stocks no matter how old it is or how far back the lithography goes. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 05/14/21 at 21:26:13 HEAT IS THE ISSUE Intel tries to hide their very large issue, whereas AMD intentionally attacks, remedy's and minimizes their much smaller heat issue and tries to fix whatever remains in a clear-cut fashion. https://liliputing.com/2021/05/minisforums-game-mini-will-be-a-compact-open-frame-gaming-pc-powered-by-amd.html http://https://liliputing.com/wp-content/uploads/2021/05/game-mini_02.jpg Simply remember that HEAT is the issue and everybody has got it ----- what does your guy does about it is the difference between them. A very short Twitter walk-around video of the unit in motion. https://twitter.com/i/status/1392438218272825353 This unit will answer the question "What could AMD do if cooled at the same levels that Intel simply requires to even be competitive"? |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 05/18/21 at 05:14:10 https://www.anandtech.com/show/16680/tiger-lake-h-performance-review/12 All embargos are lifted and the review folks are now telling us the sad sad story ....... Tested at 45 watts (with hyperclock spikes going up to 65 watts) Intel Tiger Lake H in hefty laptops gets WAY HOT very quickly and it slows way down. Tested at 45 watts (with hyperclock spikes going up to 65 watts) Intel Tiger Lake H will DRAIN your battery really fast when it is in turbo mode, but not for very long because it so severely overheats the laptop cooling system that it very very quickly and throttles down to pitiful. Intel sucks because it has to try to fake keeping up with AMD and APPLE, when it can't do that in reality. It can only "fake it" by doing a severe power sucking heat making hyper-clocking until it uses up the total thermal capacity of the system it is installed in. ================================================== EVIL FLIP SIDE IS STILL THE SAME You simply can't find enough AMD systems to actually buy one right now when you need it, so if you need a laptop ASAP you may well still pay beaucoup bucks to buy an expensive Intel room heater. ================================================= AMD BROADENS THEIR SUPPLY BASE ------ AMD will still continue to buy some center chiplets at 14nm from Global Foundries, but not exclusively any longer ---- all the old exclusivity deals with Global are all over now and AMD can freely buy from whoever actually can most quickly make up the chiplets that AMD needs right now. So, AMD still needs 10,000's more wafers full of new smaller chiplets as they can still sell whatever they can make as both Apple and AMD are clearly superior to Intel at this point in time. And in another 4-6 months both non-Intel companies will uncork brand new generations of processor improvements built at smaller lithography while Intel has this last little 10nm bit that was just reviewed as all that Intel has planned to show us for the rest of this year. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 05/19/21 at 03:37:23 What will the next 2 years bring? Samsung and TSMC will open major FAB CAMPUSES in Arizona, a big one for 6-5nm and another one for 3-2nm and below production. The lithography equipment installed will be good for 5nm, 3nm and very soon 2nm. Intel will fully stroke out on their 10nm stuff with lots more BS results that periodically "show dominance" over their competition while running WAY WAY hotter and using 2x as much power to do the same jobs. Intel will also start buying entire production runs of TSMC and Samsung built "Intel designs" in larger and larger amounts. This is problematical as Intel will have to use the TSMC tools to design these chipsets and at best the Intel final chipsets will be "same as" to AMD as far as capability and speed goes. Apple will keep coming out with new generational designs at the current best TSMC lithography that will simply amaze folks with what they can do. AMD will continue to dominate Intel on desktop and laptop until Intel actually uses their huge buying muscle power and simply buys up all of the TSMC process space that AMD depends upon. Yes, Intel will simply buy AMD out of the business by choking off their chiplet supply and Sleepy Joe Biden will just stand by snoozing and let them. The EU is a different story, and the EU will collect a lot of really big fines on Intel for doing this but this will not act effectively to stop Godzilla from going on in for the AMD kill shot. If IBM does manage to transfer 2nm gate all around technology effectively to Intel and Intel absorbs and incorporates this technology into all their newest FAB lines, then both APPLE and AMD will be permanently damaged. ARM and NVIDIA will remain an ongoing drama until final EU decisions are made. If the deal is allowed to go forward with "extra constraints" all the constrains placed on Leatherman Hwang will not make the final deal supportive of Hwang's dreams of total Industry dominance. Hwang the egotist will then drop the purchase at that point in time, refusing to pay the drop out penalties as it was the EU that changed his deal without his consent -- the EU owes the drop out money, not him. (more drama ensues) England simply does not wish to lose ARM as one of their last remaining bulwarks of computer hegemony. That and the Raspberry Pi are all they got left. China moving abruptly to take over the little island nation of Taiwan still remains out there as an abrupt total game changer. Expect American and the EU to make their planned moves as if this was gonna happen at any time now as there are detailed packets of orders already distributed to the Chinese General Command Staff for China to go do this takeover ...... Jedi Order 99, anybody? |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 05/21/21 at 12:40:02 https://liliputing.com/2021/05/leaked-amd-roadmap-points-to-6nm-rembrandt-chips-with-navi2-graphics-in-2022.html http://https://liliputing.com/wp-content/uploads/2021/05/amd-2022-700x367.jpg AMD’s Ryzen 5000 mobile chips are already some of the most powerful x86 processors available for laptops. But a leaked product roadmap suggests that next-year’s Ryzen 6000 chips could bring significant improvements. Among other things, it looks like AMD is preparing to move from a 7nm to a 6nm manufacturing process, while updating its integrated graphics architecture from Vega to Navi2. The roadmap shown above that was leaked by @Broly_X1 shows three new mainstream laptop processors on the way: Rembrandt H 45-watt processors for high-performance laptops including mobile workstations and gaming notebooks Rembrandt U 15-watt processors for premium ultrathin laptops Barcelo U 15-watt chips for lower-cost ultrathin notebooks The Rembrandt chips are 6nm processors that feature AMD Zen3+ CPU cores, Navi2 graphics, support for LPDDR5 or DDR5 memory, PCIe4, and USB4 (which may not be quite the same as Thunderbolt 4, but hopefully it’ll help fix one of the biggest gaps between current-gen Intel and AMD laptops, which is the lack of 40 Gbps USB-C ports on most AMD models). Barcelo is basically what you get if you take this year's tech and stuff it into next year’s processor. It’s a 7nm, 15-watt processor technology featuring AMD Zen 3 CPu architecture and AMD Radeon Vega graphics. AMD did the same thing when launching its Ryzen 5000 chips, with some entry-level models featuring the same CPU architecture as last year’s “Renoir” chips, but also increasing core counts to offer better overall performance. It just means that when shopping for an AMD laptop with a 15-watt processor, you may need to look closely to figure out which of the three 15-watt chips you’re getting from here on out. AMD intends to keep all of their 7nm allocations going and then add to that the newly promised 6nm allocations so as to fulfill more of the large demand that is going lacking at the moment. I suspect they may do the same with thing with 5nm allocations when they add their 5nm volume to the total AMD production allocations. Right now AMD is running fault free as all of their 7nm and 6nm versions outperform Intel's very best offerings. When this changes, I suspect AMD will drop the price of the group of "now lacking processors" and continue to offer them as a lower cost option until the channel clears out of all the old finished stock inventories. Assuming we ever manage to work the chiplet wafer supply up to getting any unsold old AMD finished stock inventories ....... ::) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 05/25/21 at 13:52:04 https://liliputing.com/2021/05/googles-1st-gen-nest-hub-now-runs-fuchsia-the-os-that-might-unify-all-of-googles-products-one-day.html Google has been developing a new operating system called Fuchsia for at least five years, but up until recently the company hasn’t actually used it on any commercially available hardware. Now 9to5Google reports that Google has begun pushing Fuchsia to a real device that people already own. A new software update rolling out to the first-gen Nest Hub smart display (it was called the Google Home Hub at launch), and that software replaces the device’s operating system with a new one based on Fuchsia. http://https://liliputing.com/wp-content/uploads/2021/05/nest-hub.jpg Here’s the basic idea behind the move – the Nest Hub user interface is separate from the underlying operating system. So Google can replace the Linux-based “Cast OS” (software that’s similar to what runs on Chromecast devices) with Fuchsia and users can still interact with their devices in pretty much the same way as they’ve been doing. Under the hood though, the operating system will rely on a new operating system that uses a custom “Zircon” kernel instead of Linux, giving Google far more control over the software ecosystem powering its device. So far Google is only rolling out Fuchsia to a single device. But if that goes well, it’s easy to imagine the company pushing Fuchsia to other Nest products and maybe then to other product categories altogether. A few years ago Bloomberg reported that Google’s roadmap for Fuchsia started with smart home products, but could eventually include smartphones and laptops. One day Fuchsia might replace Android and Chrome OS… or at least replace the backbone of those operating systems with something new. I’m not entirely convinced that Google wants to kill the brands that it’s spent the last decade or so building for those operating systems. But if eventually all of Google’s products are based on the same software, it could make it easier for developers to create apps that work across a range of platforms including phones, tablets, notebooks, smart TVs, wearables, and whatever else is next. Google started Fuchsia up to protect Google Android from a set of patent suits from Oracle and others. The suits go tossed out of court as they were baseless and quite obviously mercenary in nature, but Google had Fuchsia over half built at that time, so they now have a small quick scale-able OS system that they can use for ...... just about anything. If Microsoft tries to take over Linux or Android again (yes, nobody trusts Microsoft or Intel not to lie cheat and steal whatever they see out there) Google has another working string for their bow. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 05/30/21 at 23:41:52 https://liliputing.com/2021/05/intel-launches-core-i7-1195g7-and-core-i5-1155g7-chips-and-a-new-5g-card-for-laptops.html As expected, Intel is adding two new chips to is Tiger Lake-U family. The new Intel Core i7-1195G7 is the company’s most powerful U-series chip to date, with support for Turbo boost speeds up to 5 GHz, while the Core i5-1155G7 processor is a step up from the Core i5-1135G7 chip that currently powers dozens of laptops. Intel says more than 60 devices featuring the new chips will be available by the 2021 holiday season, including laptops from Acer, Asus, Lenovo, and MSI set to hit the streets this summer. Like other Core i5 and Core i7 Tiger Lake-U chips, the new processors are quad-core chips with support for hyperthreading and Intel Iris Xe graphics. What’s new are the top frequencies. Both of the new chips have top GPU speeds that are 500 MHz higher than the chip one step down on the ladder. And both can hit higher CPU speeds as well. Here’s the complete Tiger Lake-U family, with the new chips highlighted in bold: Name Cores / Threads TDP Base / Boost Graphics EUs Graphics Max Cache Memory i7-1195G7 4 / 8 12 – 28W 2.9 GHz / 5 GHz 96 1.4 GHz 12MB DDR4-3200 LPDDR4x-4266 i7-1185G7 4 / 8 12-28W 3 GHz / 4.8 GHz 96 1.35 GHz 12MB DDR4-3200 LPDDR4x-4266 i7-1165G7 4 / 8 12-28W 2.8 GHz / 4.7 GHz 96 1.3 GHz 12MB DDR4-3200 LPDDR4x-4266 i5-1155G7 4 / 8 12 – 28W 2.5 GHz / 4.5 GHz 80 1.35 GHz 8MB DDR4-3200 LPDDR4x-4266 i5-1135G7 4 / 8 12-28W 2.4 GHz / 4.2 GHz 80 1.3 GHz 8MB DDR4-3200 LPDDR4x-4266 Intel says the new processors aren’t just competitive with other Intel chips, but also with AMD’s best 15-watt U-series chip to date, the 8-core, 16-thread Ryzen 7 5800U. While it’s best to take benchmark results provided by a chip maker with a grain of salt since the company can cherry-pick the tests that shine its products in the best light, Intel did provide a breakdown of several gaming and content creation tests that show the Core i7-1195G7 handily outperforming the Ryzen 7 5800U. As stated, this is at a new level of hyper-clocking coming from Intel, ostensibly rated at 28 watts but actually peaking at numbers like 48-64 watts for very short peak periods. At max peak, it totally overwhelms the laptops cooling system which is then BIOS throttled until the heat dissipates. "Intel says the new processors aren’t just competitive with other Intel chips, but also AMD’s best 15-watt U-series chip to date, the 8-core, 16-thread Ryzen 7 5800U." This is Intel marketing language that tries to say that a normally running normally cooled AMD unit can't keep up with one of these hyper-clocked Intel things when it is on a hyper-clocking tear using 3 times as many watts of power. :-/ Intel must try to rack up whatever "marketing speak" claims they can make right now as the next (coming soon) two generations of AMD lithography based progress will move the goal posts out beyond the range of Intel's heavy legged place kicker. Next, realize that the larger real test houses (Anandtech and Tom's) have testing regimens in place now that have enough test duration time in them to water down the hyper clocking spikes that Intel is now so depending upon, that instead the time averaged geomean performance ratings now includes both the hyper-clocking tear and the subsequent thermal recovery slow down period so as to yield a realistic geomean averaged test result. Lastly, realize that real production runs of Intel Inside production laptop units DO NOT INCLUDE these super CPU cooler systems that Intel now recommends, yup, the ones that are absolutely needed to achieve these results that Intel advertises. Indeed, these laptop CPU super cooler systems are not even available yet to be put into ANY laptop units --- but Intel will get there eventually, supposedly. https://www.amazon.com/Intel-BXRTS2011LC-RTS2011LC-Liquid-Cooling-Kit/dp/B006588Z7G/ref=cm_cr_arp_d_product_top?ie=UTF8 http://https://images-na.ssl-images-amazon.com/images/I/31qzB4vUmYL._AC_.jpg Yep, as of June 1 this fictional Intel cooler thing is still "unavailable" so the Intel testing results that supposedly depend on this mythical cooler are just plain TOTALLY BOGUS right now ....... One questions how Intel's little custom test houses manages to run their tests when the cooling hardware needed isn't even real yet ?????? |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 06/01/21 at 09:02:53 https://liliputing.com/2021/06/amds-3d-chiplet-tech-to-bring-192mb-of-l3-cache-to-next-gen-ryzen-processors.html AMD’s 3D chiplet tech to bring 192MB of L3 cache to next-gen Ryzen processors http://https://liliputing.com/wp-content/uploads/2021/06/chiplet_01.jpg AMD is showing off technology that could dramatically increase the amount and speed of cache memory available for next-gen chips, with up to 192MB of L3 cache on upcoming AMD Ryzen desktop chips and support for bandwidth of over 2 terabytes per second. In terms of real-world performance, AMD says that will bring the same kind of performance boost we’d normally expect from a move from one chip architecture to the next (from Zen 3 to Zen 4, for example). In a demo during the company’s Computex 2021 keynote event, AMD showed two computers running the same video game. Both systems featured Ryzen 9 5900X 12-core processors running at 4 GHz and both were using the same graphics card. But one card was a standard Ryzen 9 5900X with 64MB of L3 cache, while the other was a modified version with 192 MB of L3 cache. That model achieved frame rates that were 12-percent higher. And that’s on the low end of what AMD is promising – the company says across a range of games, it’s seeing an average 15-percent boost in frames per second. AMD says the increase in memory capacity and speed comes from its 3D chiplet technology, developed with TSMC. It allows the memory to be stacked in a way that allows big increases in density and bandwidth. At 2TB/s, the L3 cache is actually faster than the L1 cache, although the latency is higher. In a chip like the Ryzen 9 5900X with three CPU core clusters, the new 3D chiplet technology allows AMD to place a stack of 64MB of L3 cache on top of each core cluster for a total of 192MB. AMD says it will begin production of its “highest end products with 3D chiplets” by the end of this year, suggesting we could see next-gen processors featuring the new technology in late 2021 or early 2022. We saw a false hyper-clocking based announcement two days ago from Intel, one that depended on very extreme hyper-clocking stuff that runs so VERY hot that the special cooler designs it requires don't even exist yet from Intel. Is this the definition of BOGUS Intel stuff or what ........ Then 2 days later AMD routinely drops their next generation of TSMC lithography right on time along with new copper bonded chiplet stacking techniques that come direct from TSMC. AMD's new products will fully support higher (chilled water) cooling levels, but will still outperform Intel using simpler fan and fin cooling systems. Rumor has it that AMD will get still more TSMC wafer allocation along with this 4-5nm move as AMD is keeping all their existing generations of wafer allocations rolling right along since Intel hasn't yet outperformed any of the existing AMD products based on the oldest 7nm TSMC technology. Intel is still tossing up 14nm units against the oldest AMD 7nm processors. Intel is throwing up hyper-clocked 10nm Intel chipsets against the very latest 4-5nm AMD chipsets. Nether of these match ups go to Intel, but mostly go to AMD on almost all benchmarks, including the Intel "special tests" that can only be run by Intel's special little test cheat houses. Intel loudly claims these "partial victories" which are pretty dubious due to Intel lying and cheating on their little specialty houses benchmarking, using specialty cooling methods that are not available in the marketplace yet to set their "record performances". AMD is expected to continue to take more and more market share from Intel since they can make up more finished chipsets now using the latest TSMC shipping generations with the 6mn lithography (and the upcoming next generations at 5nm and at 4nm will also clearly outperform Intel's 10nm products and will come with new wafer allocation increases as well). ARM and Mediatek are expected to take more and more market share from Intel using the new, better grades of Chromebook chipsets that can be built with these new smaller lithography modes. Apple is taking Intel market share with the M-1 chipset (and the M-2 when it gets here will take more again). Intel has their product lines all announced now, but none have shipped yet for the independent testing that will make lies out of Intel's dubious claims. AMD is telling all the test houses to simply go buy their vendor's assembled products and test them against Intel's assembled products using the cooling systems that the vendors are providing in those production units (by doing so AMD is injecting some reality right past Intel's latest BS specialty house testing program that uses cooling systems that cannot possibly ship inside finished laptops). AMD is confident that if cooling systems are both REAL and equivalent, their 15 watt draw units will still win out over Intel's 15-28 watt units quite handily in most finished laptop designs where Intel is sucking up 28 watts (or even more when Intel is hyper-clocking). To combat Intel's program of seriously bogus hyper-clocking AMD now allows their AMD chipsets to be run on up to 28 watts now if the laptop's cooling system is up to it. It is rumored that late this year AMD will now allow hyper-clocking spikes to 48 watts on their most modern processors if and only if the laptop builder thinks his cooling system can cool that sort of action down quickly enough for the hyper-clocking to be practical. Yes, AMD thermal loads like this may be supported now by laptop builders in some cases, but SOME REALLY STRONG cooling systems have to be really available (and fully functional) at that point in time as AMD encourages all such products to be tested with the cooling systems that are actually used in the finished units. AMD plays it straight, if you do hyperclocking your cooler system has to be up to the job as the complete laptop will be tested and sold (and reviewed by Tom's, Anandtech and the other real test houses as a packaged set). AMD has much better thermal characteristics, more cores/threads, more on chip L2 and L3 cache memory and much better set of on board graphics than Intel does. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 06/08/21 at 15:37:46 Intel is competing against TSMC ....... using tech it licensed from TSMC? Intel is in a very odd, very bad spot. TSMC has come across the pond and is putting together 3 LARGE FAB PLANTS in Arizona. Concrete is poured and walls are going up as we speak. TSMC will complete their first wave of plants this year and will have them up and running and will be building the second wave of even lower lithography fab plants started before Intel finishes their single modest first new facility. TSMC has plans going down to 2nm and Intel is planning to be building a 10-7nm plant. In no fashion can Intel compete against TSMC for general fab customers. Intel is still planning on building stuff that is 3 generations behind and the entire world will centralize on the better smaller more dense lithography that TSMC is offering. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 06/12/21 at 02:57:15 https://www.tomshardware.com/features/amd-vs-intel-cpus The newest AMD Ryzen 5000 wave and the newest Intel hyperclocked Rocket Lake responses have hit the benchmarks, and the net picture is very much more of the same old same old stuff. Each set of stair steps is a little higher than the last set of stair steps with AMD coming out on top and at least one instance of Intel hyper-clocking their latest stuff hard enough to show up on the list to claim to be "competitive" using over twice the watts of power. Biggest real world difference between AMD and Intel is that AMD is rated using the cooling system that comes in the box with the processor, while in some cases THERE IS NO COOLER SYSTEM PHYSICALLY AVAILABLE to mate up with the exaggerated claims Intel is making for their latest BS marketing Rocket Lake. With no real high heat capable cooler system available for laptops, you are actually buying off some bogus Intel specialty test house test results that will not be even close to real in the real world product you are actually purchasing. The following chart is all about GAMING (Intel's best performer segment) --- and the best performance per dollar chipset value out there is the AMD Ryzen 5 5600X PBO ar $300. http://https://cdn.mos.cms.futurecdn.net/XLpJGGjH9ahb9qkP9F3dDb-970-80.png.we Recommendation is to buy AMD if you can find it. AMD is getting snatched up world-wide at a point of sale marked up premium prices right now compared to Intel. AMD still sells direct at recommended retail off their own AMD webpages, but that source of list price supply is very spotty on the best value AMD chipsets (they are instantly sold out). AMD is making and shipping their entire allocation set of different TSMC wafer allocations worth of finished products, but this is still leaving enough unfilled demand on the table for Intel to continue to survive very nicely. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 06/15/21 at 12:09:47 https://www.eejournal.com/article/growth-everywhere-but-at-intel/ EVERYBODY grows by over 10% as the market grows, but all of these growers are taking some market share from Intel to do that growth. Up 10% or more, except for Intel of course. Intel shrinks by 4% (shrinks by 14% relative to the growing market). http://https://www.eejournal.com/wp-content/uploads/2021/05/iStock-1279138405.jpg Click on the graph to see all of it, or open it in a new tab to see it all at once. http://https://www.eejournal.com/wp-content/uploads/2021/05/61421-image1-1024x841.jpeg These numbers come to us from IC Insights, which updates them quarterly and religiously. Per company policy, they include foundries like TSMC as well as name-brand chipmakers like Intel and Qualcomm. They also include, where appropriate, sales of optoelectronic components, sensors, and discrete components, not just ICs. Only about half of the companies on the list make such components, and, for those that do, it’s only about 5% of their business. All the companies but one (ahem) saw sales increase from early 2020 to now, and all but one of those posted hefty double-digit upswings ranging from 11% (Apple) to a remarkable 90% (MediaTek), to a barely believable 93% (AMD). Intel, sadly, saw a 4% decline in sales, an $832 million drop. Samsung is predicted to lap Intel for raw size by 2023. AMD doubled in size last year, which is in line with its innovations. Intel, if they stay stumbled for two more years will not be "in charge" of anything any further. Mediatek came close to doubling in size, which was mainly due to Trump killing off their Chinese competitors with his trade war sanctions. Still, Mediatek is going to become a super power in Chromebooks which themselves are gaining in importance vs Intel & old style PCs. Intel made a lot of somewhat false BS noise last year, but still shrank as a business as 14nm is getting really really really long in the tooth now. Intel 10nm isn't lighting the world on fire either, and since that is all Intel has on deck for the future you will perhaps see Intel swing over to being more of an Automotive fab style supplier more and more as they follow the Global Foundries pathway to oblivion. Intel has been told by its board of directors to "Get off 14nm ASAP" but Intel has no ability to do that in a timely fashion. Intel 10nm and 7nm have come up repeatedly lacking compared to TSMC and Samsung's current product offerings. Specifically, Intel 7nm when it does arrive will see the world leaving 5nm and going down to 4-3-2nm from TSMC and Samsung. Once again, a day late and a generation behind ...... Wall Street has extracted a price for Intel's failure to advance, and Wall Street will do so again this fall. Intel is currently waiting for Biden's bail out money, and that Biden money will simply be too slow and too little when it finally arrives as Intel really cannot spend the large amounts of money they have available now to effectively change their company's tech posture. So far the only thing Intel is good at is hiding lots of software based AI boosters and lying about their thermals and their real test procedures. Meanwhile, so far Intel has announced plans for only one 7nm fab: the Fab 42 in Arizona. In addition, the company is going to have some 7 nm-capable capacity at its D1 facility used for development and trials (among other things). ASML still is not listing any large orders of their newest scanners slated for Intel. ASML is at "full production capacity" and is queuing all new orders with supply dates going out 2 years in the future. Intel is screwed, in other words. Dutch Government or EU "command control" of ASML would be needed before Intel could rapidly get what they needed to retool their plants. The EU wants their own production plants and ASML would supply scanners to that before they would to Intel's still not firmed up orders. All of this means Intel would be forced to build their own lithography equipment yet again --- something they are not really very good at compared to ASML. What is Intel good at? Having a supermassive ego and a denial factor that functionality puts Intel into situations like this repeatedly. Does Intel think that fudging tests and lying about EVERYTHING really is their secret superpower? The power of the BIG LIE? Intel BS "brown vapor" Marketing is their real superpower? |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 06/20/21 at 04:28:57 RUMOR TIME AMD begins taking early 5nm chiplet shipments from TSMC. Call them sample lots if you wish as they are indeed full lot runs of chiplets but they are not coming in a constant flow supply fashion. Apple is releasing some of the 5nm flow they had locked up for two years now as Apple rolls on down to 4nm and 3nm. AMD is getting part of this old Apple 5nm production flow and is in turn shipping it out as test samples to their major partners so they can get a leg up on the next future wave of AMD. AMD may do a variety of things with these new more efficient, faster chiplets. We shall see what is built with them and how the ship quantities ramp up in the real world as AMD gets their full flow of 5nm rolling out into the market place. Remember, AMD is going to continue to use all of their older lithography allocations ongoing in a full speed ahead fashion, so as to be able to take more contested market share away from Intel. This will continue until Intel finally manages to put out something that can actually beat the oldest AMD 7nm chiplet based processors ....... and Intel 14nm simply isn't up to that job. ================================================== Intel is attempting to buy out RISC-5 design house SiFive ---- "Why?" you ask. Simple, they want the most powerful of the SiFive chip designs to be their Intel "little" core as they have no appropriate little core of their own. SiFive had recently approached Intel to get Intel to run their chip designs for them, so Intel immediately turns on them and tries to do a hostile take over ....... Got any moral there for the rest of you little guys who might have some tech that is worth stealing ????? Something simple like don't dangle a hot dog in front of Godzilla when he's hungry ?????? SiFive P550 is the equivalent of the ARM A75 core and would make a good little core for Intel. The bad news for SiFive is they go from being a small industry leader to ...... nothing, as Intel will spread their people out among their design groups and promptly hack up everything that SiFive had innovated into existence right back out of existence. https://liliputing.com/2021/06/sifive-performance-p550-is-its-most-powerful-risc-v-processor-to-date.html SiFive has some production allocations at TSMC 7nm level, and Intel may want that as well. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 06/24/21 at 01:08:48 https://www.extremetech.com/computing/321796-amd-roadmap-leak-major-platform-graphics-changes-coming-in-zen-4 http://https://www.extremetech.com/wp-content/uploads/2021/04/AMD-Roadmap-Leak-640x441.jpg AMD Zen 4 Roadmap RUMOR leak --- smells true enough in general, but likely will change the details some in the actual execution. Please remember, this set of rumors was originally some future talk about Zen 4 5nm stuff. However, TSMC has just now started shipping early lots of 5nm chiplets to AMD for testing/early production runs that will make this all a reality very suddenly. Also remember the extreme AMD demand for chiplet allocations may explain some of the 6nm and 4nm variants showing on this slide that AMD may build for relatively short periods of time. AMD is VERY hungry for more chiplets and AMD will certainly use these odd allocations up and beg for more. Apple is just now releasing part of their 5nm lock so AMD's 5nm chiplet flow is just now beginning. So let's revisit Rumorland to see what we can see for AMD's next best wave of 5nm stuff. =================================================== What Zen 4, Ryzen 7000 May Bring Zen 4, Ryzen 7000 presumably launches for desktop in late 2021 or early 2022. It’ll be AMD’s first desktop platform on 5nm, its first chip to feature PCIe 5.0, and it’ll be the first time AMD has offered a graphics solution on every CPU. AMD is the company that first popularized the idea of a CPU and GPU sharing the same piece of silicon, all the way back to its acquisition of ATI back in 2006. It’s ironic, therefore, that Intel has done a better job of making baseline graphics capability available across its entire product line. Beginning with Zen 4, this changes, and RDNA2 becomes available across the entire product stack. That’s a bigger change than it might seem. AMD may have adopted chiplets for its desktop CPUs, but its APUs are firmly monolithic, including the eight-core Cezanne/Ryzen 5000 APUs available in mobile. There’s good reason for this. Despite the marketing around them, chiplets are not a unilateral positive. AMD pays a penalty in terms of die area, latency, and power consumption compared to a monolithic chip. In the desktop space, these disadvantages are small — especially considering that AMD can field a 12-core and a 16-core desktop CPU at a lower per-chip cost than it would otherwise pay. In laptops, however, AMD has chosen to stick with a more conventional design. AMD’s consoles, similarly, are monolithic architectures. Integrating a GPU into a chiplet CPU design presents certain challenges. AMD could theoretically build a monolithic GPU chiplet to sit alongside the I/O die, but the fabric requirements from this kind of arrangement would be formidable. It might work for one CPU chiplet connected to one GPU chiplet, but we doubt AMD could wire this solution into Epyc. Alternately, AMD could possibly build a GPU into every Ryzen chiplet, and distribute workloads across multiple chiplets while treating the entire array of chiplets as a contiguous graphics card. The GPU cores in each chiplet would presumably connect to the CPU cores via the L3 or potentially an L4 cache. In the diagram below, that would equate to a GPU cluster being tucked between the CCX and the “Infinity Fabric” block — separate from the CCX functionally, but still connected via the LLC. http://https://www.extremetech.com/wp-content/uploads/2021/04/Ryzen-5K-Topology-640x314.jpg We expect AMD will redesign Infinity Fabric for the Zen 4 die shrink and possibly include a new I/O die, assuming one doesn’t tip up for Zen 3+. Inserting a theoretical GPU die into the CCX topology from above doesn’t change the number of Infinity Fabric links needed across the entire chip, however, since the GPU CUs would be fully integrated into each chiplet. Every Zen 4 chiplet would contain, say, 128-256 GPU compute cores (this number is entirely theoretical and could be higher). A 12-core or 16-core Ryzen 7000 would offer commensurately more GPU cores. A 64-core Epyc with 256 cores per chiplet would offer a maximum of 2,048 GPU cores across the entire CPU. AMD has been reticent to discuss AI in much detail, but executives have told ExtremeTech that they aren’t blind to the rapid advances or long-term potential of the industry. Unlike most of its peers, including Qualcomm, Intel, and Apple, AMD hasn’t brought a low-power NPU to market or focused on adopting SIMD instruction sets that specifically improve AI performance. If AMD wants to compete in this space — and it says it does — incorporating a CU cluster into every Zen 4 chiplet would provide a guaranteed accelerator unit. Distributing a workload across multiple chiplets could also reduce hot spot formation. I suspect — again, if this leak is accurate — that this is why the RDNA2 block for Zen 4 is green on desktop but red on mobile. All of AMD’s conventional, monolithic designs, for both Vega and RDNA2, are indicated in red. We see one single, solitary green block. It shows up at the same time other rumors imply AMD will launch RDNA2 baked into every Ryzen chip. From the beginning, AMD’s big theme with Ryzen has been re-use, with the same chiplet design scaling from low-end desktop to high-end server. A Zen 4 chiplet with integrated RDNA2 hardware (or CDNA2, if AMD went that direction with Threadripper/Epyc) would provide additional processing horsepower for AI calculations by leveraging AMD’s existing IP rather than requiring a from-scratch solution. There’s no proof for this — it’s speculation on my part, and I don’t have inside information — but if the company wants to put GPUs inside chiplets, it either has to build a single unified GPU block that connects to every chiplet or it has to build a little bit of a GPU inside each chiplet. The second seems easier to scale than the former. Something to remember is that in each generation AMD builds an across the board chiplet that can run in various multiples in each level of their product line. The same chiplet design runs in the cheapest four core low power stuff as runs in the very biggest 248 core mainframe stuff. So some care is being taken right now to include a connection fabric and built in graphics capability and a built in AI capability suitable for the heaviest computer use cases into the raw AMD base chiplet design that is being done at this time. WHAT THIS ALL MEANS ----- AMD WILL AUTOMATICALLY HAVE GOOD GRAPHICS, LOTS OF MEMORY AND GOOD AI BUILT INTO EVERYTHING THEY MAKE. Nvidia will take a hit in the graphics card business again and we can predict Intel will take another abrupt dip in overall market share. THIS MOVE IS AMD'S KEY INITIATIVE THAT MAKES THEM EITHER A WINNER OR A LOSER IN THE LONG RUN VS INTEL AND ALL THE ARM BOYS |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 06/24/21 at 01:23:21 Intel makes the front page of the nightly business news for losing somewhere between 4% and 14% of their market share. Next day, Intel abruptly fires their current mainframe head dog and churns their entire internal organization structure yet again for the 4th time since they fired their old errant bean picker in chief. Intel dudes, based on your Board of Director's statements these periodic purges will continue until you guys start doing something right ....... One questions if this endless game of musical chairs is really helping Intel do any long term good stuff ?????? At some point in time lining them all up on the edge of the aqueduct and pushing every 10th one off with a spear butt becomes sorta counter productive. The guy with the spear butt keeps saying the same thing while sending his solders over the edge down the long drop ....... "Get off 14nm. Get 7nm working right." =================================================== As Huawei doubles down on making their own chipset building equipment (Huawei has no choice, Trumph's entity listing says no state of the art ASML scanners nor any other western tech for Chinese Military linked businesses). At the same time Intel acts to slowly decimate its own business groups just to get them to move off the dime technologically, all the while Huawei is also busy floundering about seeking a future pathway. ::) So perhaps it is time for us to stir some rumors about the future of electronics past 2mn. IBM and Samsung have built some normal style 2nm tech samples and some 2nm Gate All Around samples. Both of these sample runs ran into signal loss through the tiny connecting wires going to both styles of processors, making the 2nm results a lot less reliable than 3nm & 4nm results. Soon "smaller" will not be better any more. =================================================== At this same time AMD has announced they will be building distributed AI and distributing upper level graphics into the guts of every chiplet. In essence AMD will be distributing their massive GPU and AI and the localized chunks of systems memory at the localized processor chiplet level keeping it actually located inside their CPU structure. At 5nm and below, the new extra available flat real estate on the socket makes this expansion possible. Boosting the L-3 and L-4 memory up vertically on top of the top of the die also helps make more room for this functional AMD expansion. By keeping most of the action close by inside the chiplet processor itself AMD is avoiding this growing issue of large signal loss due to very small fine connecting wires. And this will actually work out to be both faster and better for AMD, as these new chipsets do not need to communicate long distances across a motherboard or through complex buss structures or through secondary communications chipset subsystems all of which stuff always act to slow down the overall data flow. Aside from the exterior inputs and post computational outputs, everything remains inside the CPU itself. Ampere and others have recognized these same small wire leakage issues in their own testing and they have developed their "full wafer sized chipset systems" to put 100% all of the computer into the resulting huge wafer sized chipset itself. AMD is doing the basically the same thing, just using their superior chiplet design in aggregate as a distributed full computer system with very close by local memory and GPU and AI distributed clusters. The AMD system is much better in making less testing scrap, as a defect in the Ampere system can scrap an entire wafer, while several flaws in the proposed AMD system result in very minimal chiplet scrap. Base upon this generic issue of 2nm signal loss due to fine wires, AMD clearly sees that they only have 2-3 future lithography shrinks left for them to utilize ....... and they also see Intel in the background kinda sorta matching AMD's current high end performance levels using much much larger and much cruder lithography running at relatively huge power draws, all the while using lots and lots of sneaky cute little AI tricks just to get the job done "fast enough". AMD can do the same sorts of AI tricks in their much faster processors, but they will approach these tricks more openly if they go there as AMD/Xylinx owns all the AI tech base patents that all these tricks depend upon. Intel does not own this tech at all, AMD owns it ---- thus the Intel sneaky "don't ever talk about it" approach Intel is using. What is also clear is the consumer need for general processors has doubled during the 2 Covid years and it will double again inside of 3 years as we recover from Covid. Intel and Global Foundries are still betting their older tech base is still actually good to go for as long as it physically lasts as the marketplace is telling them that OLDER LITHOGRAPHY IS BETTER THAN NO LITHOGRAPHY and that they can still sell all they can make at this point in time. AMD currently gets to charge a sizable premium for their best in class chipsets, and all the others must discount their prices to move their inferior designs if demand eases up any at all. Huawei is also taking this side of the bet, as they have their existing old AMD CPU licenses to produce 7nm and larger chipsets, and this is the tech level where Huawei will be building their new Chinese foundries using IBM/GF style old homemade lithography systems. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 06/28/21 at 13:50:22 https://www.tomshardware.com/features/intel-core-i9-11900K-vs-amd-ryzen-9-5900x http://https://cdn.mos.cms.futurecdn.net/HCdKq8oS59zF9Fy8A6dzrV-970-80.jpg Today finds us pitting the AMD Ryzen 9 5900X against the Intel Core i9-11900K in a battle for flagship CPU supremacy. AMD's Ryzen 5000 processors took the lead in the desktop PC from Intel's competing Comet Lake processors last year, upsetting our Best CPU for gaming recommendations and our CPU Benchmarks hierarchy. Intel's response comes in the form of its Rocket Lake processors, which dial up the power to extreme levels and bring the new Cypress Cove architecture to the company's 14nm process as Intel looks to upset AMD's potent Zen 3-powered Ryzen 5000 chips. Have no doubt; Intel is pushing the Core i9-11900K's aging 14nm silicon to the absolute limits in an attempt to steal the crown from the Ryzen 9 5900X. Unfortunately, AMD has been plagued by chip shortages due to very high demand — you simply can't find the Ryzen 9 5900X in stock for reasonable pricing due to retailer scalping — this leaves Intel an opening to capitalize on ....... BUT ONLY IF YOU CAN COOL THE BIG HOT RUNNING INTEL BEAST DOWN EFFECTIVELY. The outcome of our Ryzen 9 5900X vs Core i9-11900K battle seems pretty straightforward on the surface: The Ryzen 9 5900X wins four out of five categories, while also scoring a tie in the features section. Our CPU faceoff ends up being a five to two win in favor of the Ryzen 9 5900X, meaning that the choice should be quite clear for most enthusiasts. The Ryzen 9 5900X ultimately wins on the strength of its better blend of gaming and application performance, not to mention that it comes with much lower power consumption that ultimately results in a cooler and quieter system. And that's despite it coming with four more cores than the 11900K. AMD's successful formula has consisted of more cores, a newer architecture, and a denser 7nm node, but Intel launched the Core i9-11900K on an older, less-efficient 14nm node with fewer cores. As a result, Intel attempted to offset the reduced core count by dialing power consumption way way up to the extreme to maximize performance. That results in much higher power consumption and increased heat, so you'll need a very capable cooler and robust motherboard to unlock the best of the 11900K, all of which adds to the cost. Note that in a laptop the required Intel laptop super cooler does not exist for Intel's current designs. Once again, Intel is using a specialty test house to generate bogus laptop test results that cannot be gotten from any laptop hardware that is physically available, you know, the fictional production laptop stuff that Intel claims to be able use to get the Intel chipset to run at full tilt in their ads. In a desktop with a chilled water cooling system you might be able to get these sorts of numbers for a short period of time before BIOS throttling eventually sets in. The AMD desktop unit will make AMD's advertised numbers with the cooling system that comes in the box with the processor. Since the heat produced by AMD is so much less, a relatively standard AMD gaming laptop rig can actually do AMD's advertised results in the real laptop world if the laptop chassis and cooling system was actually built for a constant heavy duty use in a real gaming situation. Note, these thermally capable laptop systems are quite rare, actually. And these laptops are quite heavy as the heat pipe cooling systems (plural) weighs a lot. This heat divide between AMD and Intel will become even greater when 5nm AMD gets out into the real world. 5nm AMD will run both cooler and faster than what is shown in the match up above. In the same same static head to head run using same same general construction of laptops started at the same time using Apple M1 vs Intel core i7 in near identical rigs it is not uncommon for the fans in Apple 5nm M1 rigs to not even come on AT ALL until after the competing Intel i7 is already thermal throttled all the way down to its bare minimum compute speed due to simple overheating. Intel gets that hot that fast ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/02/21 at 10:25:01 More Hot Rumors https://consumer.huawei.com/ae-en/community/details/TSMC-ANNOUNCES-ITS-FIRST-3NM-AI-CHIP-CUSTOMER-NEITHER-APPLE-NOR-HUAWEI/topicId_115506/ TSMC announces roadmap plans for next 2 years At TSMC’s 26th Technical Symposium, TSMC officially confirmed that its 5nm and 6nm processes are already in mass production. In addition, the company also announced that it will release a higher version of the 5nm process next year. Furthermore, TSMC officially confirmed that more advanced 2nm, 3nm, and 4nm is in development. As for the 4nm process, it is an ultimate improvement of 5nm. However, the 3nm process is the natural successor to the 5nm process. In terms of technical indicators, the risk production of the 3nm process (N3) will take place next year. The mass production will commence in 2022. Compared with 5nm, 3nm will reduce power consumption by 25-30% and improve performance by 10-15%. The 4nm (N4) risk production is also scheduled to take place next year. Just like the 3nm process, 4nm mass production will start in early 2022. For TSMC N5 customers, the transition to N4 will be very smooth, which means that the tape-out cost will be greatly reduced. https://www.tomshardware.com/news/apple-intel-set-to-use-tsmc-3nm-node-for-2023-products-report Confirmation of 3nm rumors Apple and Intel will be the first to adopt Taiwan Semiconductor Manufacturing Co.'s (TSMC) N3 (3nm) fabrication process when the contract maker of chips deploys in late 2022, Nikkei Asia news agency reported on Friday. Intel is projected to use the technology to make CPUs for PCs and servers, whereas Apple is expected to use the node for its system-on-chips aimed at client devices. Apple and Intel are currently 'testing their chip designs' produced using TSMC's N3 process, according to a Nikkei report that cites several people with knowledge of the matter. Given that N3 is about to 'officially' enter risk production mode in the coming weeks and assuming that the information from the news agency is correct, we can speculate that Apple and Intel have already finalized their N3 CPUs and SoCs. Though it is unclear whether they have functional silicon. Commercial production of these chips is set to start in the second half of 2022. Intel is said to be preparing at least two products made using TSMC's N3 node: one aimed at notebooks, another for servers. There are no details about these processors at present, but Intel has already accidentally confirmed TSMC-made Xeon SoCs aimed at various niche markets. Intel confirmed months ago that it is working with TSMC on its 2023 products, but refrained from revealing any details. TSMC 6nm, 5nm and 4nm (Mediatek) are shipping volume production runs right now. 3nm starts at Apple in limited production starting in just a few months. AMD picks up all the 5nm production that Apple lays down and will start picking up most of the 4nm and 3nm as they become available in volume also as Apple moves on down to a smaller lithography. Intel has supposedly contracted a few simpler processors (some small and some large) to be run at 3nm to get their foot in the water. Intel needs some 3nm bragging rights very very badly right now but jumping way ahead to try to run neck and neck with Apple on 3nm processes that were specifically developed and are being specifically tuned for Apple sounds like a fairly high risk endeavor for Intel. Sounds shaky, in other words. Looks like Intel's Board has put a pistol to the heads of existing Intel management team and has very seriously told them that the Board will pull the trigger as many times as it takes to get Intel completely off 14nm ASAP. One schedule slip or market reversal = one trigger pull. Intel is also trying at the same time to say their in house 7nm will be almost as good as TSMC's 3nm, something that sounds a lot more like some more of the typical Intel BS marketing misinformation. This was BS back when Intel first said it about 5nm, and it smells worse when said about 3nm. Remember, TSMC will cut Intel off if they keep on acting like a "spoiler" or "completely bogus acting" customer, so Intel should be somewhat mindful of their behavior. Remember, Intel is now a direct foundry competitor for TSMC, not just a simple customer. As such, Intel is not altogether welcome at TSMC right now, having failed to utilize some contracted allocations when they became available in the very recent past. Next, AMD has already contracted some sizable allocations for 3nm TSMC already, but AMD will use up these allocations up and then try to phase into any additional 3nm chiplet allocations they can get their hands on while keeping up all the applicable earlier lithography nodes of chiplets still flowing at full stream (as AMD is still very very hungry for more wafers of chiplets). Remember, AMD is pulling market share away from Intel just as fast as they can get the flow of appropriate chiplets to do it with. AMD is a signed up money providing ground floor player in all of the new USA and new EU TSMC plants that are under construction right now. Intel is not a ground floor "paying player", not yet anyway. Intel is acting like they are "entitled" right now due to Biden using them as front men for his new tech initiative, but Intel should also note that Biden has not given them their megabucks as he cannot legally do that apart from Congress passing appropriations and getting them approved by both houses. TSMC's recent actions to resolve the current automotive supply issues renders Biden's grandiose give away plans somewhat moot ...... SUMMARY AMD 7nm beats Intel 14nm and most early Intel 10nm AMD 6nm beats all Intel 10nm AMD 5nm and 4nm will likely beat early Intel 7nm (whenever, if ever it arrives) AMD 3nm and 2nm will likely be equivalent to Intel 3nm and 2nm, with the winner having the best integrated thermal design benefits. Will AMD's new designs with distributed local graphics and distributed local memory and distributed local AI carry the day for AMD? We shall see ....... =================================================== https://liliputing.com/2021/07/qualcomm-plans-to-launch-a-laptop-processor-that-rivals-apples-by-mid-2022.htm Qualcomm intends to make a M-2 competitor chipset using their NUVIA tech which will take some additional share away from Intel as Intel owns that general market segment right now but is rapidly losing it to the Apple M-1 and M-2 chipsets at this time. http://https://liliputing.com/wp-content/uploads/2021/07/qualcomm-700x394.jpg In an interview with Reuters, Qualcomm CEO Cristiano Amon said that he thinks his company will deliver a processor that’s competitive with Apple’s by mid-2022 when the NUVIA purchase finalizes. This entails additional market share losses to Intel as well as to Apple proper. There aren’t a lot of details explaining what to expect from that chip in terms of features or performance. But if you’re wondering how Qualcomm plans to catch up to Apple, it’s with the help of some former Apple chip designers that Qualcomm has hired. Earlier this year Qualcomm acquired a startup called NUVIA for $1.4 billion. That company, which was formed by a team of chip designers with experience working at Google, ARM, Broadcom, AMD, and Apple. NUVIA brings new, state of the art processor and AI tech to Qualcomm. CLARIFICATION Qualcomm is going to be using their new NUVIA designs to do this, as existing ARM designs are simply not as good as the new NUVIA processor designs promise to be. ARM is expecting to counter against this NUVIA Initiative with some brand new ARM processor designs and and ARM will work very hard to beat NUVIA to market, thus preempting the Qualcomm initiative. NVIDIA taking over ARM has hit up against the Chinese regulators now, and the Chinese are simply not going to respond to it at all, thus killing the take over when the final due date comes and goes. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/05/21 at 15:25:28 THIS IS NOT NUVIA, THIS IS CHISEL BASED https://liliputing.com/2021/07/open-source-xiangshan-risc-v-processor-could-eventually-challenge-arm-cotex-a76.html CHINA UNIVERSITY DEVELOPS MORE POWERFUL RISC-5 CHIPSETS MODERN CHISEL HARDWARE LANGUAGE USED YIELDING 50% SMALLER & FASTER CODE BASE http://https://liliputing.com/wp-content/uploads/2021/07/xiangshan-logo-768x599.jpeg XiangShan has been developed as an open source project with a BSD-like Mulan PSL v2 license. Since its inception in June of 2021 contributors have submitted more than 50,000 lines of code and published 400 documents. One of the more interesting features of XiangShan is that its code is written in the Chisel hardware description language. Its creators say that resulted in a codebase that’s 1/5 the size it would be if it had been written in the older Verilog language. All of these new players will take some more relatively small amounts of market share (and image) away from Intel ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/07/21 at 11:22:12 https://liliputing.com/2021/07/intels-unpopular-lakefield-chips-reach-end-of-life-a-year-after-launch.html Intel dumps embarrassing "very first" Big - Little chipsets out of the lineup. The Intel Core i3-L13G4 and Core i5-L16G7 processors are five-core chips that combine four low-power, energy-efficient CPU cores based on Intel Tremont architecture (which is basically a version of Intel’s Atom technology) with a single Sunny Cove CPU core (using the same architecture as Intel’s 10th-gen Ice Lake chips). Theoretically this was meant to allow PC makers to build thin and light computers that offered long battery life while delivering performance on par with what you’d expect from an entry-level machine with an Intel Core processor. In practice, only a few devices ever shipped with Lakefield processors, including the Samsung Galaxy Book S and Lenovo ThinkPad X1 Fold. The general consensus from reviewers was that you could get better performance from a device with an older Intel Core m3-8100Y processor (like a $630 Microsoft Surface Go 2) than you’d get from these devices, which sold for $1000+ and $2500+ at launch, respectively. So in that sense, it’s not particularly surprising to see that Microsoft has issued a Product Change Notification indicating that it’s discontinuing its only Lakefield processors. What is a little surprising is just how quickly the company is doing that. At the time Lakefield came out and got tested no one could see any advantage to Intel's Big Little Lakefield family. It all felt like an ARM copying "me too" exercise that was actually carried out in a very poor, nonsensical fashion. Intel has now excised the old team leader for Lakefield, with inference that he did something in error, but instead he did exactly what his bosses at that time told him to do. Intel is also coming across as sort of a nasty place to be working, right now ......... on top of lying repeatedly to the public about their processors performance. =================================================== Some new information comes to help explain the abrupt Intel Lakefield dump off and its replacement by Intel Alderlake processors. https://liliputing.com/2021/07/intel-alder-lake-mobile-chip-lineup-leaked-up-to-16-cores-with-bigsmall-designs.html Alder Lake processors will use Intel’s “Foveros” 3D stacking technology to package high-performance CPU cores based on “Golden Cove” architecture on the same chip as energy-efficient, Intel Atom-based “Gracemont” CPU cores. Similar to ARM’s big.LITTLE technology, this allows chips to leverage the right CPU cores for the task at hand. Need more horsepower to complete a job? The big cores kick in and deliver a burst of performance. Running less resource-intensive jobs on your computer? The lower-power Atom cores can probably handle things, extending your laptop or tablet’s battery life. Alder Lake isn’t Intel’s first crack at this sort of heterogenous computing architecture. The company launched an Intel Lakefield processor with one high-performance “Sunny Cove” core and four energy-efficient “Tremont” cores last year. But while it proved the 3D stacking technology worked, the chip delivered underwhelming performance and Intel has discontinued it just a year after launch. If the leaked product slide is accurate, it looks like Intel is being much more ambitious this year, with a set of chips arranged across six different product segments, half of which are entirely new (marked by an asterisk below): *Intel M5 – 5W-7W chips for tablets with Lakefield-like 1 big + 4 small core designs and 48 or 64 GPU execution units Intel U9 – 9W – 15W chips for ultra-thin laptops including 2 big + 4 or 8 small cores and 80eu or 96 eu graphics (there may also be a 1 big + 4 small + 48eu chip in this range) Intel U15 – 12W/15W/20W mainstream laptop chips with similar properties to the U9, but higher power consumption (and presumably higher clock speeds) *Intel U28 – 20W – 28W “performance” chips with either 4 big + 8 small or 6 big + 8 small cores and 96eu graphics Intel H45 – 35W – 45W “thin enthusiast” chips for gaming laptops and workstations with 4 big + 8 small or 6 big + 8 small CPU cores + 96eu graphics *Intel H55 – 45-55 watt processors with 8 big and 8 small cores + 32eu graphics, likely because these chips are designed for “muscle laptops” where they’ll be paired with discrete graphics (there may also be a 4 big + 8 small core version) Intel groups the M5 and U9 chips under the Alder Lake-M family, while the U15, U28, and H45 are considered Alder Lake-P chips. The H55 chips are are part of the Alder Lake-S family. While Intel usually restricts “S” series chips to desktops, it seems like the H55 processors are part of the company’s mobile lineup, despite being BGA socketed chips. So far these chipsets look like a continuing execution of a family of FAIRLY EXPENSIVE Intel FOVOROS chipsets, same idea that got last year's first generation of the same abruptly dumped off last week. Good luck determining how powerful the new Intel 14nm processors really are or how many watts of power they actually draw while doing whatever it is that they are doing ....... it sort of depends on whatever it you are using your machine for, in other words. Dynamic switching and dynamic throttling on the fly say that Intel processors are going to be a constant soup of "change up" depending on exactly how the laptop maker configures the machine this week and exactly how you are using it right now. And this configuration and set up can be changed by broad brush strokes by a nightly over the internet update from Intel's servers !!!!????? Holy built-in cover-up, Batman !!!! |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/10/21 at 23:28:55 Intel seems to be driving towards "the completely unknowable" just as fast as they can get there simply so they can hide their overall lack of progress better. AMD still keeps on lowering their power consumption while speeding up their processing. They do this by lithography shrinks and better and better internal chip structure. Inside the next year, AMD will be bringing more motherboard functions into the socketed chipset itself, using inside the CPU processor traces instead of a separate complicated I/O bus that is managed by a separate chipset cluster. AMD intends to have major motherboard elements built into the chiplets themselves that will aggregate together to be the GPU, be the AI and be the systems memory. The more complex the actual chipset that you buy, the better your aggregated GPU and etc. will be, naturally. Power consumption will be much lower, and communications within the combined chipset will be as simple and as fast as it is currently possible to get, as the distances are low and the only I/O will be a final answer or a finalized graphics output of some sort. TSMC 5nm lithography can have over 20 layers per chiplet, and each chiplet layer will be larger in area by 25% compared to older lithography nodes. This is a lot of extra room to work with. AI acceleration will simply be state of the art. Xylinx co-ownership means that some of the 20 chiplet layers may contain Xylinx FPGA tech that allows the chiplet structure to be modified on the fly in ways never seen before. Pre-built modifications can be stored on the machine itself and the "custom built for the task at hand aggregate on chipset AI" can be built up from these stored modifications quite quickly. If someone figures out a new systems attack, AMD can make it go away by changing their Xylinx FPGA CPU structure to make it impossible for the exploit to work any more. AMD will begin doing this next year, while Intel is still dicking around with their "me too" Big Little stuff that doesn't work very well. =================================================== Intel likes to run full out at HUGE energy consumption levels that are just plain flat out at CPU destruct heat levels. TSMC is going into 3nm pre-production right now with both Intel and Apple as the initial customers. TSMC is dealing with large hot spots inside these new Intel test chips and dealing with the overall Intel overheating in a new very forthright manner. TSMC is filing patents on how to route water cooling passages into the distinct layers of their 14-24 layer 5nm and 3nm chipsets. The size of the passages are quite small as are the sizes of the heat exchangers and the tubing/piping on this new type of machinery is tiny, but the effects on processor durability and in the positive effect of keeping the processing running up at the faster levels cannot be denied. https://techunwrapped.com/tsmc-is-studying-using-liquid-cooling-inside-the-chips/#Integrated_liquid_cooling_by_TSMC http://https://techunwrapped.com/wp-content/uploads/2021/07/TSMC-is-studying-using-liquid-cooling-inside-the-chips-780x470.jpg Click and read a bit --- this new technology is being pushed by TSMC and Intel. This type of liquid cooling can work on a laptop as the passages and piping are tiny. Heat pipe systems already in use in the more powerful laptops can support this form of very focused liquid cooling. WHAT DOES THIS MEAN? INTEL CAN STILL SAY THEY REMAIN "COMPETITIVE" WITH AMD AND APPLE BY USING THESE TRICKS, BUT AT A COST OF LARGER BATTERIES AND MUCH LOWER BATTERY LIFE. APPLE apparently does not need the trick, as they don't even run the fan on their units except sporadically. Apple's designs are very power and heat friendly compared to Intel. AMD is going to need to use the processor level cooling trick also, when they actually put all the graphics and AI and memory functionality distributed deep into their combined function CPU chiplets. Current draw on AMD processors will jump up from 105 watts to 170 watts ---- with most of the up level in heat coming from the now included somewhat hotter running distributed graphics that used to reside in a different, separately cooled graphics card. TSMC's in-processor liquid cooling will be potentially part of every brand of state of the art processor, including AMD's very best. Please note that 107o up to 170o still means the AMD distributed graphics system will be much more heat efficient and cooler running than NVIDIA or Intel graphics, by a factor of two at least. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/14/21 at 14:44:58 https://www.sammobile.com/news/volkswagen-cars-using-samsung-exynos-auto-chips/ Volkswagen’s new cars are using Samsung’s 4mn Exynos Auto chips All modern cars use semiconductor chips for various functions, including vehicle controls, internet connectivity, in-vehicle entertainment, navigation, and safety. A few years ago, Samsung expanded into the growing automobile industry with its Exynos Auto chipset, and the company has been supplying chips to Audi since then. It is now being reported that even Volkswagen, the world’s biggest auto brand, has started using Samsung’s chips. According to a new report from South Korea, Samsung started supplying Exynos Auto chips to Volkswagen in early 2021. The chip is reportedly used for multimedia playback, navigation, and vehicle status control. The Exynos Auto V9 chipset is made using Samsung Foundry’s 8nm FinFET fabrication process. It features eight ARM Cortex-A76 CPU cores, an ARM Mali-G76 MP18 GPU, and an integrated NPU (Neural Processing Unit). Its DSP supports up to six displays, while its ISP supports up to 12 on-vehicle cameras and 4K 120fps videos. Samsung is planning to expand its Exynos Auto chipset lineup in the future, given that more and more cars will depend on chips for data processing, image and video processing, internet connectivity, voice recognition, and other tasks. Samsung recently unveiled its first ISOCELL image sensor that is specifically designed for automobiles. The company also makes OLED screens, LEDs, and batteries for electric cars, and they are used by BMW and a lot of other auto brands. Since more cars are bound to switch to full-fledged operating systems such as Android Automotive, QNX, and Apple’s OS (for its rumored smart car), sales of semiconductor chipsets are bound to rise in the near future, and Samsung wants to be fully prepared to cash in. The company is already working with Hyundai, Genesis, and Tesla on various projects. Intel has promised a 2 year plan to meet automotive needs by building new Intel facilities local in Germany and in Japan. Intel's grandiose plans are $20 Billion dollars short and two years way way too late. Samsung, predicted to take the #2 worldwide leadership position in building chipsets away from Intel inside this year, has rung in with Volkswagen already as their main chipset supplier ---- deal is done, finito, completed right now. TSMC is already the world's leading foundry, so this does not change either with Samsung booting Intel away from #2 position. Question is, can Intel hold on to #3 or is it fated to lose that as well? Done deal, past tense, water over the dam --- Exynos is coming inside Volkswagens starting RIGHT NOW. Volkswagen is the world's largest car maker, so Samsung now has the Automotive Leadership in chip making all sewed up already. Samsung and TSMC have been tasked by world governments to end the Automotive Chipset Shortage ASAP and Samsung and TSMC has a scheduling plan to do that within the next quarter that does not involve building new buildings nor costing several 20 billion dollar programs like Intel wishes the EU and Japan to do. Samsung and TSMC will simply short cell phone production while supplying Automotive Chipsets as a government mandated priority. This will be all be done inside 4 months. Samsung swinging over to full on automotive opens up new business potentials for Mediatek and Xiaomi, and Samsung is actually helping Mediatek to move into the larger position in cell phones to stave off Xiaomi's meteoric rise. Samsung/Mediatek/Xiaomi is a powerhouse trio that is pushing Qualcomm aside quite quickly in cell phones. You will start to see more ARM computer chipsets and ARM automotive chipsets starting RIGHT NOW. Samsung's consortium is moving in on Intel, taking the new markets away from both Intel and Qualcomm. ::) Intel is a joke and is being treated as such by the EU and Japan. Biden's putting Intel in charge of the USA's computer recovery plan is a grave mistake as all Intel is doing now is the same old same old Intel BS self-delusions that are now being pushed over on others as Intel grabs for all the free cash money they can get. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/15/21 at 01:07:41 https://www.fudzilla.com/news/pc-hardware/53214-chipzilla-unlikely-to-catch-up-in-semiconductor-fabrication http://https://www.fudzilla.com/media/k2/items/cache/94450839300597d598cb23722397ef86_L.jpg Economic academics agree Intel squandered its lead trying to impress their own Intel shareholders A couple of economic academics have penned a rather interesting paper claiming that Intel will not ever catch up to TSMC and SEC in semiconductor fabrication. William Lazonick and Matt Hopkins, writing at Institute for New Economic Thinking have been looking at the numbers to work out why Intel fell behind TSMC and SEC in semiconductor fabrication. Their theory is that Intel is engaged in two types of competition, one with companies like TSMC and SEC in cutting-edge fabrication technology and the other within Intel itself between innovation and financialisation. The Asian companies have governance structures that vaccinate them from an economic virus known as "maximising shareholder value" (MSV). Intel caught the virus over two decades ago. As we shall see, with the sudden appointment of Gelsinger as CEO this past winter, Intel sent out a weak signal that it recognizes that it is sicker than a dog and needs to quarantine. In the years 2011-2015, Intel was in the running, along with TSMC and SEC, to be the fabricator of the iPhone, iPad, and iPod chips that Apple designed. While Intel spent $50 billion on P&E and $53b. on R&D over those five years, it also handed over more than $36 billion to shareholders in stock buybacks and $22 billion in cash dividends, which together absorbed 102 per cent of Intel's net income. From 2016 through 2020, Intel spent $67 billion. on P&E and $66 billion on R&D, but distributed almost $27 billion as dividends and another $45 billion as buybacks. Intel's ample dividends have provided an income yield to shareholders for, as the name says, holding Intel shares. In contrast, the funds spent on buybacks have rewarded share sellers, including senior Intel executives with their stock-based pay, for executing well-timed sales of their Intel shares to realise gains from buyback-manipulated stock prices. If Intel had not been so busy trying to impress Wall Street and its own shareholders it might have lavished more money on R&D and equipment and not have a TSMC problem, Lazonick and Hopkins say. https://www.ineteconomics.org/perspectives/blog/how-intel-financialized-and-lost-leadership-in-semiconductor-fabrication Detailed analysis as to Intel's long term errors and why Intel isn't coming back into leadership, ever. Our policy recommendation for the Biden administration is simple: As a condition for giving the U.S. semiconductor industry $50 billion in infrastructure assistance, put a ban on SIA members doing stock buybacks as open-market repurchases. That legislation can then be a first step in Congress rescinding the Securities and Exchange Commission’s Rule 10b-18—corporate America’s license to loot.[26] With a critically important company like Intel focused on innovation rather than financialization, the United States can get back to the business of building a world-class semiconductor-fabrication industry – one that leads rather than lags advances in technology. Dopey Joe Biden can't even follow this sort of thought train from beginning to end, much less try to do something about it. Heck, Dopey Joe can't even read his teleprompter right, which leads to yet more of his infamous whispering episodes. Unless Congress changes the law, big business will always chose to reward "bad behaviors" simply because they are GREEDY. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/15/21 at 20:55:48 https://seekingalpha.com/news/3715822-intel-in-talks-to-buy-globalfoundries-for-30b-wsj https://www.wsj.com/articles/intel-is-in-talks-to-buy-globalfoundries-for-about-30-billion-11626387704?redirect=amp This is a hoot ....... Biden has put Intel in charge of renovating American chip building. Biden is a weak bulb mentally, and that was all he could think to do, go give snow shovels worth of cash to Intel BS people who will waste it just like they have wasted their own money over the last 10 years. Intel is not even in the ASML 2 year waiting line to buy any new state of the art lithography equipment. That ship sailed over a year ago while Bob Swan was still in charge of Intel. Intel has now burned all its bridges now trying to "consort" with TSMC and Samsung, both of the more modern larger chipmakers are now treating Intel's feeble attempts to get back in to the foundry business like they should treat any brand new weak kneed budding competitor. TSMC will now treat Intel like a pay up front high risk customer, which is what Intel is in reality. Samsung really isn't working with Intel at all at the moment, their relationship was strained to breaking from past "projects" that went south due to inaction on Intel's part. Intel has some current deadlines from the Biden Administration that they are failing to even make a dent into. No progress, no more Biden money, in other words. So, Intel is now trying to secretly buy the corpse of Global Foundries from the Arab conglomerate that currently owns it. Why? To meet Biden's deadlines, Intel will GO ABSOLUTELY WASTE ~30 billion~ American Tax Dollars on something that is just about totally worthless for any future oriented uses. But it is American soil based "production" which is what Intel was charged by Biden to go increase. This is a bullshite trick by Intel, as the capacity was there last year under Global Foundries and it does not constitute a singe wafer of new capacity that wasn't there a year ago. Saying Global was Arab owned and was "non-American production" is really really really BS stretching things a whole whole lot .......... Intel's specialty -- brown vaporous BS Saying "it is good for automotive uses" is pretty much wrong headed too, as Volkswagen for example doesn't want to use Global at all, they want to use brand new 4nm tech direct from Samsung. Not any 15 year old moldy 20nm tech from Global Foundry. Remember, VW is the world's largest car maker now a days ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/16/21 at 22:18:49 So, how is the cell phone marketplace shaking out with Huawei being "mostly gone"? https://www.canalys.com/newsroom/global-smartphone-market-q2-2021 http://https://canalys-com-public-prod.s3.eu-west-2.amazonaws.com/static/press_release/images/12131563132021Q2_Smartphone_Prelim%20-%205.jpg “Xiaomi is growing its overseas business rapidly,” said Canalys Research Manager Ben Stanton. “For example, its shipments increased more than 300% in Latin America, 150% Africa and 50% in Western Europe. And as it grows, it evolves. It is now transforming its business model from challenger to incumbent, with initiatives such as channel partner consolidation and more careful management of older stock in the open market. It is still largely skewed toward the mass market, however, and compared with Samsung and Apple, its average selling price is around 40% and 75% cheaper respectively. So a major priority for Xiaomi this year is to grow sales of its high-end devices, such as the Mi 11 Ultra. But it will be a tough battle, with Oppo and Vivo sharing the same objective, and both willing to spend big on above-the-line marketing to build their brands in a way that Xiaomi is not. All vendors are fighting hard to secure component supply amid global shortages, but Xiaomi already has its sights set on the next prize: displacing Samsung to become the world’s largest vendor.” Russia is investing in RISC-V processors for cell phones and computers. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/17/21 at 20:42:16 My Goodness, Intel has really messed this little ranking up by renaming all their processes and their part families without actually improving anything. This whole chart has just now (a week after it was written) become completely trashed by a billowing cloud of brown Intel BS vapor blowing over it ......... https://www.tomshardware.com/reviews/cpu-hierarchy,4312.html And Tom's Hardware had just retested everything using Intel and MS's latest released software tweeks (tweeks that make the hidden Intel AI work a little better, in essence). Rankings didn't change anything position wise, but the stair steps went up a little bit. There is a presumption here that Intel doesn't have another AI trick up its sleeve, or yet another "missile lake" of some sort to spring on us. This is the final ranking before AMD releases their next lithography shift and announces all their new products. Intel has to wait for new TSMC produced (internally cooled processor) 3nm products to have any chance of a top rated product. 1-2 years from now, huh? It is sad that Intel depends on a breakthrough in processor cooling to try to become competitive again ........ It is also noted that most of the software tweeks mentioned up post actually were apparently SPECIFICALLY DONE TO CHEAT ON A SPECIFIC BENCHMARK TEST with the goal being to show at least one Intel chipset product up in the top two or three in every test that is commonly used. None of which changes a GEOMEAN ranking much at all ....... Tom's did a good job making a cheat proof set of benchmarks to stop Intel's little mickey mouse games. However, Intel is still hooting & hollering and claiming this and claiming that, all of which would be a lot more real if their stuff didn't throttle itself to death inside the first 2 minutes of real use. If Intel had a better cooling system then their results would be a lot more real than they are right now ....... The crushing pace of AMD real advancements continues, wave after wave of improvements are already contracted at TSMC and will be be produced late this year and early next year. Intel can bump up AMD's release schedule by putting out something compelling, but the odds of this happening go down month by month. AMD is putting out 5nm sample products to their OEM partners so everybody can get ready for the next waves of AMD stuff ...... Reminder, because of AMD prodding Intel with AMD progress, computers in general are 3 times faster than they used to be ........ Both vendors are still needed to supply the Very Large consumer demand for new PCs. AMD and Intel can both sell all that they can make right now., but demand is going to drop off soon due to Covid ending (but back to school is picking up to counteract this relief). http://https://cdn.mos.cms.futurecdn.net/aYQAe9Mp8jrntAeQPCfa7n-970-80.png.we |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/26/21 at 14:59:59 Today, Intel tosses it all up in the air, calling their old stuff 10nm & 7nm & 4nm & 3nm and saying they are as accurate in doing this as Samsung is with their current chip lithography destinations. This is blatantly false, and when the Intel news embargo ends the counter blasts from the computer press will begin. ::) Some of the positive outcomes of this blasting controversy may be the forced development of an impartial universal nomenclature that is based on current draw per CPU thread. The hard fact is that Intel is losing currently on all fronts and desperately needs to reset the discussion into new terms they can lie better at. I have said previously that Intel is simply driving very hard to become "unknowable" so as to appear to be competitive when they certainly are NOT COMPETITIVE AT ALL. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/26/21 at 15:09:54 http://https://cdn.mos.cms.futurecdn.net/hT7SXey3SXehbusK36epKY-970-80.jpg Intel tosses it all up in the air, saying their old 14nm stuff is now 10nm Superfin (of two different varieties no less with some being called 7nm now) and renaming their failed 7nm stuff as 4nm & 3nm and 2nm and less. Intel is saying they are being as accurate in doing that as Samsung is with their current chip lithography destinations. BULLSHITE ALSO NOTE: Intel is using "embargoed information" discussion shut downs to stop people from calling Intel out for their rank utter BS nonsense. ::) Some of the positive outcomes of this blasting current controversy may actually be development of an impartial industry wide universal nomenclature that is based on current draw per data unit measured at post throttle levels, a metric that has some real meaning and is easily measured and reported. Intel will fight this change hard as they really suck at it, greatly. Until then, expect Intel to lie a whole lot and shoot out massive clouds of brown BS vapor to try to confuse all issues on all fronts while still claiming to be better than everybody else. The hard fact is that Intel is losing BADLY currently on all fronts and Intel desperately needs to reset the discussion into some sort of new terms they can lie more convincingly about. I have said last week that Intel is simply driving very hard to become "unknowable" so as to better appear to be competitive when they certainly are NOT COMPETITIVE AT ALL. RECOMMENDATION: BUY AMD AS YOU CAN STILL GET REPORTED MEASUREMENTS FROM AMD THAT STILL HAVE REAL MEANING, produced using the stock AMD heat sinks that AMD puts right in the retail box with the processor. Some specious Intel 10nm new stuff requires some form of super liquid cooler to get their Intel advertised results, a cooler that simply does not exist right now in reality. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/27/21 at 01:00:48 https://www.youtube.com/watch?v=0PD7IJgbuWs This is a calm relatively factual relatively unbiased YouTube video attempt to analyze what Intel is doing, with this analysis being done by a lesser level tech reporter. Pay close attention to the fact that all Intel 14nm and 10nm changes are immediately tossed under the bus according to this reviewer with the simple notation that Intel must actually execute this plan before he (the reviewer) will try to talk about it any further. Nobody wants to play "what if games" with a company like Intel that never brings what it promises to market at the same very broken level as Intel has done lately. Simply put, it is all BS until Intel actually manages to DO something to change something. Until then, it is rated as yet more Intel brown vapor BS and intentional confusion. Intel has supposedly hinted that it intends to buy a whole lot of brand new ASML machines at the real 3-2 nm range, but Intel simply hasn't ordered nor paid for any of these new scanners yet. The time to place these orders has come and gone just recently ....... There is a two year long waiting line to buy these things and Intel isn't in that line yet. Intel has no large state of the art "build it for me" orders lined up at TSMC at this time at any existing TSMC fab either. However, after embargo ends tomorrow Intel will begin magically naming their old stuff down at these reduced numerical levels according to the "numbers that have no meaning" approach that Intel is currently taking. Needless to say, for PC processors BUY AMD instead of Intel ........ buy the current modern state of the art and know what you are buying, for Heaven's sake. Also, if Chromebooking, buy a Mediatek Dimensity 1300 instead of Intel for a lower cost better performing chipset that has known operating characteristics....... I CANNOT RECOMMEND BUYING ANYTHING FROM INTEL SIMPLY BECAUSE THE NAMING OF THE PROCESSORS HAS NO MEANING ANY MORE AND CAN BE FLIPPED AROUND FAST LIKE A FAST WAGGING PUPPY TAIL. Intel is no longer a trustworthy source for your electronics since LYING TO EVERYBODY is apparently their NEW secret superpower. Next, Intel has now ruptured the deal they had going with Samsung, split it WIDE open at the waistline so to speak. Busted a trust gut wide open Intel did ....... So now Intel is claiming to be close to QUALCOMM now and will be supposedly building Qualcomm's PC lite processors instead of Samsung or TSMC building them as has been in years past. For Qualcomm's sake, I hope this is not true. Don't believe Intel when they say they can build an ARM like SOC at low lithography levels, Intel never has been able to do this before now, ever, now have they? I think Qualcomm is simply due a hard sharp lesson on why nobody deals with Intel in any kind of close, reciprocal relationship ---- not for long, anyway. Intel screws their business partners over regularly, leaving them holding the bills for the high Intel scrap rates and other inherently high Intel project monetary costs. Here is another tough one, Apple has recently doubled down on their 5nm and 3nm orders at TSMC with Apple intending to take more x86 market share using this mass of processors. Yeah, Apple, the ones that paid TSMC to develop the new equipment to build this new stuff way way back then --- this development leaves very little state of the art processing space currently available for Intel as AMD was second in line to Apple at TSMC and TSMC does not willingly default on existing published agreements. AMD will get some super modern chipsets as per that agreement. Apple and AMD may get very tight on these TSMC new chip supplies, but TSMC will take care of these long term TSMC partners. Why? Both major players also list Samsung as alternate suppliers and if the two major players both have to pick up on these alternate options at Samsung that leaves NOTHING for Qualcomm and Intel except "So Solly, Charlie". The general effects of the worldwide chip shortage just grind on and on and on finer and finer, don't they? |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/28/21 at 07:34:04 The computer industry roils some more due to TSMC's limited state of the art capacity being simply unable to permit all of their old customers to all grow at the same time. Some customers will have to wait a bit as other customers are going first. These guys will lose market share to the customers who can get some TSMC chipsets. EXAMPLE: Apple still has most of TSMC 5nm and 3nm locked up with massive massive orders. Apple states they intend to use these orders to take market share away from other players. Intel and Samsung seem to be separating again on their on again off again relationship. Qualcomm and TSMC and Samsung seem to be sorting each other out in light of Intel's announcement that Intel will be building Qualcomm's PC chipsets going forward. NOTE: This Intel announcement is not verified by Qualcomm at this time ....... Mediatek has announced two new serious Dimensity 1300 heavy duty chipsets that can do a light PC duty very well, and would be simply kickass as Chromebook chipsets. These will be built at the TSMC 4nm level. First lots are now built and shipped to machine builders, so this is not just a rumor. Qualcomm is simply not competitive to Taiwan based Mediatek when taking the long view any longer. This takes down Microsoft's PC lite efforts and may well be what is firing off the shift to Intel by Qualcomm, IF this indeed happening. Qualcomm and Microsoft are primarily aimed at the American market, which is no longer the big growth leader for buying lots of tech stuff any longer. The really large phone based growth markets are now covered by Mediatek and Xiaomi as Huawei has been cut off from use by all American technologies by the Trump and Biden administrations. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/28/21 at 18:48:15 https://9to5mac.com/2021/07/28/tsmc-2nm-production-set-for-2023/ https://gizmodo.com/tsmc-will-start-making-2nm-chips-as-intel-tries-to-catc-1847381091 https://asia.nikkei.com/Business/Tech/Semiconductors/Intel-sets-2025-goal-to-regain-chip-crown-from-TSMC-and-Samsung More announcements from Intel, with Intel speaking out of their arse "for TSMC" about Intel's plans to do this and that "to beat TSMC" using chipsets built at TSMC until Intel can build their own new process lines in America proper using Biden's money. Big Issue is that Intel's plans take an additional two years just to get into 2nm production compared to TSMC's plans that were announced SEPARATELY today. TSMC has plenty of time to over-run Intel's plans using not yet invented progress in chip building technology. Apple will insist on using this new tech starting next year ........ Furthermore, the Taiwanese government itself today gave gov. approval to TSMC to expedite both their 3nm and their 2nm facilities, actually moving them forward by one full year. Land is there, facilities hookup is there, slab is poured already. Walls going up soon. Intel has just likely ended their chance to get their Intel products run at TSMC's new facilities by acting such a fool about the whole thing. Or worse yet, TSMC may hold Intel to their illicit, unauthorized "for TSMC" announcement and just delay giving Intel any 3nm or 2nm production allocations for one additional full year's worth of delay as per Intel's illicit announcement dated today. :D Yeah baby, teach them stupid suckers to talk your confidential top secret stuff out of school ...... Intel, you are either a trustworthy customer or a blabbermouth competitor Fab, so pick one ----- soon TSMC will stop telling you things and letting you play the fool and embarrassing them by you announcing TSMC's secrets that do not belong to you as you have just done. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 07/30/21 at 17:48:12 https://www.digitaltrends.com/computing/everything-amd-new-am5-socket-reveals-about-zen-4/ http://https://static.tweaktown.com/news/7/9/79551_05_check-out-this-mockup-of-amds-next-gen-zen-4-raphael-am5-cpu.jpg Which would you rather have? Early mock ups of the next gen AMD and Intel chipset match up. And yes, because of AMD's tighter lithography and AMD's use of chiplets and "complete inside the chipset communications wiring" the AM5 socketed AMD unit is indeed smaller and simpler to connect up to a motherboard. Intel requires many additional components on the back of their chipset and has to have nearly twice as many CPU connections and twice as many traces to their motherboard (leading to still more sub-component set ups and required support items). Intel simply costs a whole lot more to implement and their motherboard costs reflect this much higher cost posture. http://https://pbs.twimg.com/media/E7e1Y0-WEAEAJAC?format=jpg&name=small http://https://pbs.twimg.com/media/E7e1a8DWQAA3lh3?format=jpg&name=240x240 =================================================== Market Segregation by price Premium end buyers don't want Intel right now. AMD is supplying that entire upper end market at some serious premium+ price points. AMD is getting the very best bang for their TSMC chiplet allocations by supplying mainframe and high end & workstations first, then letting Intel supply all the bottom end chipsets using their lies and older lithography when the monthly AMD TSMC chiplet allocations run dry. Sad thing is Intel wants to lie about just about everything now, about what they are supplying and the lithography that is used inside it...... and Intel is charging a lot of ++ money ++ for a lot of relabeled older processors right now ........ Buy Intel and YOU ABSOLUTELY DON'T KNOW ABSOLUTELY WHAT THE HECK YOU ARE ACTUALLY GETTING FOR YOUR HARD EARNED MONEY RIGHT NOW. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 08/01/21 at 16:02:06 We got us a new inflection point, courtesy of Mediatek. The AMD Ryzen 3 chipset's throughput specs has been overcome by the 4nm 8 core Mediatek Dimensiity 1200 and 1300T chipsets. Yes, upper end cheap ARM based Mediatek tablet and Chromebook chipsets have exceeded the existing lower end Ryzen 3 performance. We will see more of this sort of low end overlap action in the next several years as the Mediatek ARM based chipsets get better but still sell for far far less than the existing Ryzen 3 chipsets. AMD has paid zero attention to this low end area as they do not have chiplet wafer allocations right now to use their chiplets to support Ryzen 3 properly right now. BTW, Chromebooks are going to continue to get cheaper and better, too ----- the low end Intel Chromebook chipsets are many, but are quite weak comparatively. ================================================= Looks like the Chinese regulators are simply never going to act on the NVIDIA attempt to buy ARM, and by doing so will quietly exercise their veto power. This leaves ARM open to doing an IPO to raise some buy out capital, or by agreeing to a joint customer buy out from their many many customers. This sounds familiar to us all, because ARM Holdings was nothing but an earlier customer consortium buy out of ACORN, the original ARM chipset organization. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 08/02/21 at 10:23:36 https://www.wired.com/story/google-tensor-pixel-6-pro/ https://liliputing.com/2021/08/lilbits-google-pixel-6-with-tensor-processor-steam-deck-and-upgrading-the-gpd-win-max.html https://media.wired.com/clips/61044e27f5d5b47a960fd94b/360p/pass/Google%2520Tensor%2520(Video).mp4 it is a video, click on it Google built a tech called Tensor Flo to do their heavy duty learned AI processing, which they called Tensor Processing. https://en.wikipedia.org/wiki/Tensor_Processing_Unit#Fourth_generation_TPU Google used these processors in Google rack farms to do all their expansion over the last 4 years. Once allowed to learn a task, each tensor processor was good for 20-90+ terra ops per second. (not mega, not giga, terra ops per second) A very very fast, very very LOTS & LOTS of processing power. TPUv1 TPUv2 TPUv3 TPUv4 Date Introduced 2016 2017 2018 2021 Process Node 28 nm 16 nm 16 nm 7 nm Die Size (mm2) 331 < 625 < 700 < 400 On chip memory (MiB) 28 32 32 144 Clock Speed (MHz) 700 700 940 1050 Memory (GB) 8GB DDR3 16GB HBM 32GB HBM +0 98 TDP(W) 75 280 450 175 TOPS 23 45 90 ? Then Google learned how to plug this into the net to access this sort of terra level rack farm processing power per second in real time using a cell phone. Suddenly foreign languages instantly self-translated to English, and other amazing sorts of things started to happen on Pixel phones. Google has begun just now making and shipping early samples of their own "in the phone" Tensor processing chipsets for Pixel cell phones and tablets, and it is called Tensor silicon. It isn't plane Jane ARM, it isn't x86, it isn't Apple, it isn't RISC-V. It is Tensor (something new) Gen 1 Tensor will be being analyzed by the computer press very shortly. Google is not pumping this Tensor stuff up wildly like Intel does their latest and greatest, they simply say it is a different way of designing a processor, it WORKS differently and it can be revolutionary in both speed and in greater graphic accuracy and greater data crunching usefulness. We will learn a lot more about Tensor this fall when it ships in the new Google Pixel phones ....... The Google Pixel 6 and Pixel 6 Pro are coming this fall and they’ll feature a brand new design with a large camera bar across the back of the phones and upgraded camera hardware including new sensors and lenses. But the biggest chance, Google says, is under the hood. Google’s Pixel 6 series phones will be the first to ship with a new system-on-a-chip called “Tensor.” It’s a Google-designed system-on-a-chip with enhanced AI capabilities that will enable better photography, improved video (with HDR applied to every single frame), and other features for advanced voice processing for dictation, translation, and captions. Google hasn’t announced pricing yet, but the company other details on social media, through its website, and with journalists. http://https://liliputing.com/wp-content/uploads/2021/08/google-p6_01-700x412.jpg While Google has a track record with making its phones stand out through AI-assisted software (Pixel phones have long had a reputation for taking stellar photos, largely due to Google’s software-based camera features), I’ll be curious to see how much of a difference Tensor really makes in real-world performance… and whether any apps that aren’t made by Google will tap into its features. https://www.youtube.com/watch?v=7hEPj13PUGc This dude went to Google's release presentation and he went to the workshops that followed. Tensor is a 4nm Samsung produced piece of custom Google silicon that has some Samsung Exynos ARM tech inside it. He makes some good points and tells us what to watch for as far as new Tensor functions. https://www.youtube.com/watch?v=l9SHUSP_LJ8&list=RDCMUCRjSO-juFtngAeJGJRMdIZw&start_radio=1&rv=l9SHUSP_LJ8&t=3 Gary from Gary Explains really isn't impressed with Tensor as we currently understand it, mainly because Google Tensor isn't being transparent for much at all at this point in time. In short, Gary Can't Explain Tensor and he doesn't care for that particular fact for very much ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 08/03/21 at 08:39:12 More Minor news also happening today ......... Intel has just lost their "largest chipmaker in the world" crown to Samsung. This was predicted to take place next year, but it came early as more customers have actively dumped off Intel lately than were anticipated. Samsung in turn has lost its world's biggest cell phone supplier crown, this crown goes to Xiaomi as they roll into Huawei's old markets. I have covered the "why" ongoing over the last year, but Apple, Google, and lots of oriental companies (like Mediatek who is starting to take over Chromebook space) have all added up to some quite large market share losses for Intel. Or, let me state it this way. The compute world has grown 14% just this past quarter and Intel hasn't moved upwards for years now --- Intel actually shrank 4% last month in real terms (-14% in relative terms to all of Intel's competitors). All of this caught up with Intel's industry ranking this month. Going forward, Intel ships "mostly" old 14nm tech Core i3 and i5 now days. Mediatek has started taking over the Core i3 from the bottom up, killing the various Atoms, and the other lower market segments (Chromebooks and below). So Intel is simply in a world of low end hurt right now. Intel will make a lot of noise about shipping "more product than ever before" and they really are numerically doing that, but this is mostly due to AMD basically abandoning the bottom of the i3 range while Intel is still busy shipping out large numbers of technically moldy low end chipsets from their warehouse stocks. But with Mediatek coming on strong in these same ranges this situation will not last. Intel will have to put out better low end stuff or Mediatek will take that business. Intel losing out to Mediatek is completely absurd --- Intel has products above this range to move down into a "compete and win" range with Mediatek. Issue is the selling price to compete with Mediatek is too low for a competitive Intel upper something to make any real money competing against Mediatek. Intel will have to sell at a loss to keep a market segment that they really do not want and that they cannot possibly keep. Next gen AMD Ryzen 3 (AMD 5000 6000) is apparently going to overlap up into Intel Core i5 range by the end of these next AMD product generations and AMD 5000-6000 could wind up being a good bit cheaper too (this being mostly dependent on the TSMC wafer supply's abundance, actually.) AMD can take whatever market share that they have the wafer supply to handle. Apple currently has the TSMC major wafer supply flowing to them, so Apple will take the most short term market share away from Intel instead of AMD. Now do you begin to understand WHY Intel is driving so hard to increase all levels of confusion about exactly what the heck they are actually selling and how good it actually is ????? Intel is fighting 3 battles at once right now using obsolete lithography and a high cost position. If they focus on one competitor the other two rip out a mouthful and swallow it while Intel contests with the third ....... ........ blood and squid ink in the water, boys and girls, lots & lots of murky Intel squid ink and more and more blood ........ You do realize this will signal Intel's Board of Directors to take out their little small caliber pistols and inject 90 grains of hot lead into another minor Intel manager's cranium, right? Somebody has got to pay for this huge huge loss of Intel face after all ......... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 08/06/21 at 23:00:35 Big picture time ....... TSMC has already started enough new production facilities to get the world back into the technically current chipsets supply levels that it needs. Much of the shortage seen was people cancelling orders with TSMC due to Covid and finding out the hard way that TSMC had lots of other customers in line waiting to take that allocation space. Companies had forfeited their place in a multi-year long line, in other words. Right now TSMC is making up a huge glut of chipsets up for the entire world, chips that will be paid for up front then shipped when ready ......... and the various companies involved will still have to learn how to schedule their chipset production lots in a constant and timely flow fashion and how not to panic all the time. TSMC and Samsung will completely FINISH their USA plants and run them at rate for a year before Intel finishes the first of their two new plants. If Intel does like Intel generally does, they will dump off on their too late construction efforts and leave someone else (Biden Administration) holding the construction bag. Intel should not be allowed to buy Global Foundry with Biden's bail out money, Global Foundry was an obsolete run down turd last year and it smells even worse today. =================================================== OK, so Intel picked up a moderate bump in sales in the i3 segment, but lost share big-time everywhere else. Then Mediatek moved in on that low end i3 segment that Intel was doing well at and is killing it completely. AMD is picking up some additional wafer supply from TSMC and is using that wafer supply to take more market share away from Intel. Apple is getting TONS of wafers of M1 and M2 out of TSMC and Apple is taking a whole market direction shift out of Intel's arse at this point in time. Several different players are sifting through Intel's remaining market share, looking for any good bits. Apple, AMD, Mediatek, Amazon (AWS ARM Graviton Servers) ........ not much is going good for Intel at the moment. INTEL IS STILL MAKING MONEY, but the market share loss will hurt again this fall and again next year during Intel fiscal reporting. Can Intel sell everything they make still? Yes, but at a reduced price and reduced gross profit margin. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 08/12/21 at 20:57:20 https://www.techradar.com/news/intel-locks-down-all-remaining-tsmc-3nm-production-capacity-boxing-out-amd-and-apple http://https://cdn.mos.cms.futurecdn.net/Y2hD2oYgBKnBWLE4Kr2RjP-970-80.jpg.we Intel locks down all remaining TSMC 3nm production capacity, boxing out AMD and Apple By John Loeffler - Computing Staff Writer about 9 hours ago Intel's 3nm orders include a new GPU and three server chips NOTE: THIS WAS A HOAX Intel may have pulled a fast one on its industry rivals by buying up a majority of the 3nm node capacity at TSMC in order to fabricate its new GPU and a number of server chips, something that could inhibit AMD and Apple from ramping up production of their own next-gen chips in 2022. Production with the 3nm node is expected to start in Q2 2022, with mass production expected to start in mid 2022, according to Wccftech. Production capacity would reach about 4,000 in May of next year, with mass production capacity ramping up to 10,000 wafers a month. Intel, unlike AMD or Apple, has its own fabrication plants that it uses for most of its chip production, though it has struggled in recent years to hit its own development roadmap targets. None of the 3nm node process orders will be for its consumer market processors, namely Raptor Lake, which is expected to launch in mid-to-late 2022. Instead, reports indicate that the product lines in the order will coming out of its graphics and server units, specifically a new GPU and three new server processors, most likely next-gen Xeon processors meant for data centers. Analysis: Is the Intel Iris Xe graphics card finally making it to production? We don't know much about any of these chips yet, though the GPU could be the long-awaited Intel discrete graphics card based on the company's Iris graphics processor. The Intel Iris Xe graphics card has been in the works for a long time now, but we've yet to ever really see it beyond some prototypes and presentation material. If Intel is making a substantial investment in its graphics unit however – and eating up a substantial chunk of TSMC's 3nm node capacity in the process – then we certainly hope this is an indication that Intel's discrete graphics card is on its way to customers. While we're not expecting it to immediately dethrone AMD and Nvidia in the graphics card space, turning things into a three-way fight rather than a head-to-head matchup will push the three companies to innovate even more. This can only be good for gamers and other PC enthusiasts in the end, assuming we're ever able to get our hands on any of these graphics cards in the first place. Note the words "locked down the remaining 3nm capacity" applies to only one TSMC facility. TSMC has multiple 3nm facilities being built at this time. This implies Apple and AMD got their first pick of production allocations and had left some on the table for Intel to scoop up and Intel took all of what remained. Intel is doing some graphics cards and three workstation chipsets with this allocation ---- this is not a broad across the board Intel wide implementation. Intel will do those on their own home made lithographic equipment when they finally get that upgraded equipment ready. Intel is also known for dropping out on their long term commitments on far off TSMC production lots when these far off TSMC production lots get overcome by events. There are also some known nagging issues with TSMC 3nm long fin finFET --- it is suspected to be the very last of the finFET levels before ribbon based gate all around takes over completely industry wide at 2nm. Samsung (supported by IBM) is the best at gate all around that exists right now, having a working 2nm gate all around production equipment and a good product design system all at the same time. AMD and Samsung are tight at the moment, sharing graphics tech and production space with each other. It will be good to see some REALLY NEW tech coming out of Intel that is actually up at the forefront of things again. It will spur everyone else to do better and to do it faster. =================================================== A counter rumor has surfaced saying that Intel did NOT lock down any 3nm capacity at any single TSMC plant, but only placed some TSMC 3nm orders with a selective start and stop range. This counter rumor has been verified twice now, so the original Intel sourced rumor about Intel getting any form of lock on TSMC production is now considered a pure Intel brown vapor intentional hoax. This sort of shite coming from a competing Fab (Intel) this should be actionable by EU regulatory bodies that are aware and actually doing their jobs ...... https://www.hardwaretimes.com/amd-reportedly-books-5nm-and-3nm-capacity-with-tsmc-for-2022-2023/ http://https://www.hardwaretimes.com/wp-content/uploads/2019/12/wikichip_tsmc_logic_node_q2_2019-1024x481.png?is-pending-load=1 https://www.hardwaretimes.com/apple-to-be-tmscs-only-3nm-client-in-2022-followed-by-amd-no-3nm-chips-for-intel-till-2023-report/ |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 08/16/21 at 21:28:26 Dis-associative reality and "wishful thinking" is a danger in the computer press right now. Intel is seen by this wishful thinking Intel booster crew as still being in spitting range of the very best AMD has put together to date. Intel is behind, but fairly close on everything except temperature and current draw (which Intel sucks at, BTW). But Intel is seen as "fairly close" and almost competitive by the friendly computer press who blindly believes whatever lies Intel tells them in their press kits. This will change when Ryzen 6000 comes out next year. Between now and then there will be 2 large advancements made by AMD. The first advancement will be structural, plugging in a large layer of memory tied direct to the chiplets through brass VIA tech as patented by TSMC. This will be a ~20%~ advancement by itself. The next will be a lithography shrink to 5nm or 6nm, both sizes will be used ongoing until 3nm rolls out in Q3 2022. This will be good for another up to 30% improvement. The last item is a new TSMC "inside the chipset high performance liquid cooling system" which is under development that will make both Intel and AMD processors work better by removing much of the current overheating problems at the source. Intel will attempt to keep this trick "Intel specific" in popular reporting, but it will be a generic TSMC technology which will wind up being offered to everybody and it will be widely used as it has become needed by everybody. AMD will need the "inside the chipset high performance liquid cooling system" too when they start putting graphics cores and AI cores distributed inside every chiplet they make. AMD will rack up layer after layer of real progress and AMD will make real processing improvements while moving to 3nm by stages. Ditto for Apple and Huawei who will move down the TSMC pathway one step at a time. Intel will attempt to counter with still higher watts of current draw and some improved cooling and a few new AI tricks. And lots and lots of Intel lies and brown vaporous marketing BS ........ Intel says it is going to go to use 3nm TSMC on a limited number of specific processors, which will be designed and built using TSMC supplied processor design tools and technology. HOWEVER, Intel will stay stuck overall at native Intel 14nm and at 10nm until this actually happens. Intel has indeed already renamed their lithography levels on everything starting right now, but this basic "no real progress" truth still remains no matter what Intel calls their stuff this week. Intel will continue to lose real market share to both Apple and to AMD and to Huawei in the mean time ........ Can Intel make a come back? Heck yes, if they have the money and the time to do so. And I think Pat Gelsinger has the motivation and the management skills to run Intel going out into the future. The biggest issue Intel has right now is that they have so many new direct competitors that are doing things so very well at the moment. While Intel focuses on the current main competitor, two more smaller competitors take out a bite of hamburger and swallow it. Next biggest risk for Intel is Joe Biden --- Dopey Joe hasn't got the breadth of mind to even understand what he is doing much less what he asks folks like Gelsinger to run out and go do for him with our technical base. Next, Joe Biden has just in essence told the Chinese they can come in and invade Taiwan whenever they want to, the recent Afghanistan complete bungling has sent this message out to the entire world --- American Military isn't able to keep up with Dopey Joe as he fumbles around so ineptly. Lastly, Intel has gone off into their own brown marketing BS cloud breathing so deeply now they don't seem to understand the difference between everyday reality and Intel's own wishful thinking propaganda any further. Suddenly having all their old stuff be instantly "7nm and below" is a leap too far, and Intel's tweeking of their internal financial metrics to be in alignment with their new tweeked marketing reality will shortly run afoul of Wall Street in a very big way. Lying about money does not fly on Wall Street, the Street has no sense of humor at all. Joe Biden is self-delusional, if he exports this mental illness into Gelsinger's already loosely reality based Intel management group or gets them involved in some sort of marketing fur ball in Europe, well, then both Joe Biden and Intel could fly their planes into the ground still wing to wing with Intel still following Joe Biden's shaky lead. Dis-associative reality is a real danger for Joe Biden's close friends right now. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 08/19/21 at 19:41:29 Intel has a pattern that they follow when AMD announces and then ships something good. Go out really really far in the future and announce something to be equally good or better. Go out so far in the future that folks will have forgotten the announced details way before you fail to do them. Announce a bogus new standard of some sort to make AMD's version of something good "not fit the new standard". Publish bogus test results from a little custom test house claiming Intel does better, but do it on a thing you have no plans of ever releasing so you can't get quickly caught out in your lies. Feed false information to some tame reviewers, but only the ones that slavishly trust your press packets and will just reflexively report it as fact. Let them destroy their own credibility and then deny whatever you told them in private "confidential" meetings ever took place. But do still remember to print their bogus information on all your boxes and cartons ........ I am not alone in seeing a pattern of BS from Intel marketing. https://semiwiki.com/semiconductor-manufacturers/302408-tsmc-wafer-wars-intel-versus-apple/ The big fake news last week came from a report out of China stating that TSMC won a big Intel order for 3nm wafers. We have been talking about this for some time on SemiWiki so this is nothing new. Unfortunately, the article mentioned wafer and delivery date estimates that are unconfirmed and from what I know, completely out of line. From there the media created a frenzy pitting Intel against Apple and AMD in a war of wafers as a desperate attempt to get cheap clicks: https://www.msn.com/en-us/news/technology/intel-locks-down-all-remaining-tsmc-3nm-production-capacity-boxing-out-amd-and-apple/ar-AANfRXX https://wccftech.com/intel-grabs-majority-tsmc-3nm-capacity-4-server-graphics-chips-production-q2-2022/ https://hothardware.com/news/intel-tsmc-3nm-chip-capacity-next-gen-xeon-gpu-demand https://www.gizmochina.com/2021/08/16/apple-secure-majority-tsmc-3nm-over-intel/ And now we have the wanna be influencers on Seeking Alpha and LinkedIn repeating this false narrative ad nauseum on purpose just to generate some clicks. HERE ARE THE FACTS: I first heard word of Intel having the TSMC N3 PDK in the first part of 2020 which was a bit of a surprise. Intel is a long time TSMC customer due to acquisitions but not for native Intel products. I confirmed it with multiple sources inside the ecosystem and started writing about it shortly thereafter. What I was told later is that Bob Swan signed the N3 deal with TSMC due to the delays in Intel 10nm and 7nm to motivate Intel manufacturing to get those processes out as planned due to Intel's various internal problems. TSMC then increased their own CAPEX to build the additional N3 capacity required to satisfy the signed Intel wafer agreement. To be clear, wafer agreements are signed 2-3 years before the chip makes it into HVM and TSMC can build fabs faster than that so there will be no N3 shortages for anyone who signed a wafer agreement (apple, AMD, NVIDIA, QCOM, etc…) and actually paid the agreement costs. If they need more chips than what they signed up for, which happens, there may be shortages. If you need more, you sign a new agreement for the extra wafers. This is how TSMC and the foundry business works. It’s all about the wafer agreements. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 08/21/21 at 14:14:24 Why is AMD still on Pcie4 instead of Pcie5 like Intel claims to be? No Pcie5 drives and stuff are out there yet, there are no shipped "fully enabled" Pcie5 Intel motherboards from Intel yet to our knowledge either. Intel claims it is all eminent, but not quite here yet. Why should AMD rush into blowing bunches of money at this gnarled up Pcie5 situation until Pcie5 works itself out into a "simple reality", something that generally happens when Pcie6 gets announced sometimes next year. The gnarly parts of Pcie5 will get overtaken by Pcie6 and get a new chance to actually make it into reality. AMD for one likes to be able to re-use their old motherboards for as long as possible. Intel will have required 2-3 new motherboards be purchased during this same time span before AMD goes over to Pcie5 when it finally becomes a firm real working standard. Intel prefers for their Intel fanboys to have to buy lots & lots of new stuff with every new Intel generation whether it works well or not. "Forced motherboard churn" is an Intel thing after all. AMD prefers to let you drop in your new processor in your old socket and reuse all your old stuff for as long as possible ...... So AMD will stay at Pcie4 until a motherboard shift to Pcie5 is REQUIRED by something real, then they will announce the organized Pcie5 changeover complete with all the bits and pieces it requires at some logical point. One would think the AM5 socket processor introduction could be such a point, but we are not the ones in charge of AMD's buying of all these bits and pieces. AMD will hold off on Pcie5 a bit longer until the shortage situation gets better on Pcie5 parts and all the bits and pieces are freely available at a reasonable cost. These pieces may actually wind up being Pcie6 pieces if this gnarled up mess holds on long enough. Yup, there is even some talk about combining Pcie5 and Pcie6 as the new standard comes out this year as Pcie5 really hasn't actually done very much in the industry yet, with a lot of key players holding off on Pcie5 due to Covid waves, shutdowns and and various chip shortages subside. http://https://cdn.vox-cdn.com/uploads/chorus_asset/file/22741735/Screen_Shot_2021_07_26_at_5.33.38_PM.png Qualcomm is "excited" for what Intel promises, and Qualcomm will get Intel to quote on their new phone and small laptop stuff going forward in addition to TSMC and Samsung. That's all the commitment Intel has from Qualcomm, really, the chance to quote once Intel actually builds the plants. Then, remember, Qualcomm will keep their TSMC wafer allocation going and just add Intel as a secondary source until such time Intel can supply the entire amount at a winning price point (a never ever will take place event). It is funny that the tech that gets Qualcomm excited actually belongs to TSMC and to AMD instead belonging to Qualcomm or Intel --- funny about that little factoid, isn't it? Intel's higher pricing isn't going to win Intel very much of anything especially if TSMC actually WANTS to retain that business ........ =================================================== Intel has taken a contract from the US Military to build a plant to make sub-1.8 nm sized stuff for them in year 2025. Intel says they can do this with help from IBM as IBM is currently at the state of the art in ribbon FET technology. IBM has made the up the pre-production 2nm ribbon FET that Samsung is getting ready to produce some chipsets off of. Lots of stuff to work through on this one and Intel does not have the tech base for this super fine EUV that Samsung has developed with ASML and ARM over the years. Not even IBM has any real production level tech at this level, they currently only have a single ASML sampling machine that is capable of running the parts we are talking about, and that only at a very slow rate. Intel thinks they can modify the stuff they already own to do this job, in essence making up their own processes like they have done in years past ....... smoking their own brown vapor marketing dope again I see. I wish Intel lots & lots of luck. DARPA is their primary contracting customer for this 120 Billion dollar project and if you defraud the American Military they got a deluxe cell jest a waiting for you in Guantanamo Bay, Cuba. NOTE: Please don't try to build this complex in the deep desert south west ---- Arizona, New Mexico and Texas have all totally run out of water this summer. Chip plants need LOTS & LOTS of water ongoing all year long ....... and once that water is used and reused at the plant a time or two it picks up too much heavy metal contaminates to be sent downstream to be mixed in with somebody downstream's drinking water or veggie plant water. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 08/24/21 at 20:15:01 Samsung and the other memory chip makers are beginning to ship the 2x denser memory format that they have all been working on. Look to see the amount of memory considered to be "standard" or "normal" suddenly double or quadruple again on all your common devices as this memory becomes "what is normally available". Until then, it will be a cost adder item. =================================================== The amounts of memory installed in off lease Dell Precision items does indeed keep surging up and up. I have been eyeballing the used Dell Precision systems rolling back in off lease for relatively cheap and it is rare for them to carry less than 8 gigs of systems memory. 12 gigs is seen fairly commonly in the 6 core systems with 8 gigs in the L3 cache and one stick of 4 gigs in the normal memory slots (or 2 sticks of 2 gigs in the older units). I keep reminding myself these Dell Precision systems come out of the mainframe side of Intel, not the PC side. Once again I ogle the watts of power required to run these things, then I put it away again as a "bad idea" for me right now. ::) ================================================== https://liliputing.com/2021/08/imagination-plans-return-to-the-cpu-market-with-risc-v-chips.html Imagination, the old graphics company that Apple used to use, has come out with an announcement of a new CPU system that is not MIPS or ARM or Apple. The new Imagination system is RISC-V based. Several others are coming out with their very first RISC-V processors by the end of this year as well. Most of these are companies that are rolling off MIPS or rolling off of ARM, going over to RISC-V while intending to keep their costs at a minimum (and intending to not get entangled in the ARM/NVIDIA buy-out approval quagmire. Samsung is also putting out a Exynos chipset to compete with the Apple M-1 which is being used at the tablet/laptop level now. Samsung is sharing graphics tech with AMD on this effort. All of these arrangements will take continue to take market share from Intel, taking it from the Atom levels and the Core i3 levels. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 08/28/21 at 06:09:08 New Wrinkle from Wintel Intel is passing some gas about some really great results from their latest big little processors. But only if you use the AI based Intel Processor Manager software that comes preinstalled on your most modern Wintel 11 machine. But only if you use Intel Turbo Boost, which overclocks your processor intentionally until it uses up your thermal envelop 100%, then goes it into BIOS driven slow down mode for 15 minutes while recovering. Make sure you collect your data point before the big slow down happens ....... But only if you use the most modern Windows 11 that has the latest versions of Intel drivers built into it. But IF and only if your whole new PC machine is TPM certified by Intel and by Microsoft. IF not, then don't expect it to boot Win 11 or nor to work any at all ........... Folks, I sense a big new wave of juicy new Dell Precision trade-in full sized desktop machines heading our way to be converted to Linux Mint Mate machines. ;D |
|
Title: Re: AMD & others --- post AMD dominance in 2020 Post by Eegore on 08/30/21 at 08:30:13 Where did you get this information? If I need to make sure all four of your described parameters are met I need to hire another staff member to make sure it's done when I get my new PCs. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 08/30/21 at 16:13:29 http://https://images.anandtech.com/doci/16495/EXPLODED%20VIEW%20HORIZONTAL%202000px_678x452.jpg Yep, chiplets https://www.anandtech.com/show/16495/intel-rocket-lake-14nm-review-11900k-11700k-11600k https://www.anandtech.com/show/16857/overclocking-with-intel-rocket-lake-four-core-i911900k-binned-and-analyzed http://https://images.anandtech.com/doci/16857/Intel%20Core%20i9-11900K%20Overclocking%20Feature_678x452.jpg Eegore, all of this only applies if you are trying to beat AMD processor's current out of the box stock cooling performance by carefully stacking the deck so you can get the very best Intel test results, yep, the ones you want to report on the outside of your boxes. I do not think any of it applies to your real world situation for very much. Only Intel will need to do this right now, and only for a machine or two at one of their little custom test houses. Then Intel will splash their claims from the exercise all over all their boxes and roll on to counter the next real AMD improvement wave. Eegore will buy his machines already loaded with Windows 11 and all these goodies should have been done for him (if they apply to his machine, that is). Gotta have one of these, too. Important stuff, heavy duty cooling ...... http://https://images.anandtech.com/doci/16857/CoolerMaster%20Masterliquid%20ML360%20Cryo%20CPU%20Cooler_575px.jpg ......... do you get the point that unless you run out and buy a brand new state of the art Wintel machine with all these bits and pieces in it YOU ARE NOT GOING TO GET ALL OF THESE INTEL BENEFITS. 16 chiplets, requires a new motherboard and a new super cooler ----- all of these are needed just so Intel can claim to be competitive with AMD again. Wow. It must really suck to be Intel right now. Next, do you really need all that gee wiz stuff anyway? Here is a very cheap 4 core 12 gigs of system memory $49 used machine that can run Linux Mint Mate just fine. I would just have to drop my little SATA hard drive into it and my Mint would just work like it does now. The machine is a local pick up in Santa Fe Springs, California which is the main rub from my point of view. I would pay the cost of the machine to ship it to North Carolina. https://www.ebay.com/itm/124765229218?hash=item1d0c9650a2%3Ag%3ArCYAAOSw5WZgwQL6&LH_BIN=1 http://https://i.ebayimg.com/images/g/Ck0AAOSwlWNgwQL~/s-l64.jpg http://https://i.ebayimg.com/images/g/LREAAOSwQplgwQL7/s-l64.jpg |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 08/31/21 at 13:13:30 RUMORS ....... Samsung and TSMC are hitting some serious issues with 3nm gate all around technology while trying to run it at full production rates. Face it, 3nm gate all around is a brand new technology and they are just learning how to do it at all at volume right now ....... It is completely normal to have these sorts of issues on a brand new never been done before lithography process. Apple has been told by TSMC they will have to hold at 4nm FinFET pending resolution of the 3nm gate all around production scaling issues that TSMC has been seeing. This delay has been confirmed and re-communicated by Apple public released information. Apple has had the 5nm and 4nm allocations to weather this delay as this was all part of that tremendous 5nm run up that Apple just completed. Intel's early TSMC lots at 3nm is Shite Out of Luck, in other words. For now, anyway ..... AMD will be delayed by this as well but AMD had 5nm and 4nm firm production plans already, plans which can buffer them over just like Apple is doing. AMD is still moving forward with their existing plans to roll down to 5nm and 4nm starting late this year. We are seeing Intel forced to regroup around their old in house 10nm and their old in house 14nm (sorry, I can't keep track of what they are actually calling their old lithography this week, sorry). Speaking of being Intel and being sorry, here is more information coming from Google on starting to make their own Chromebook ARM based processors which will begin shipping next year. I mean, seriously, if Mediatek can do it, Google can do it better. Ditto for Qualcomm. https://liliputing.com/2021/09/google-may-be-designing-chromebook-processors-too.html Now Nikkei Asia reports that Google doesn’t plan to stop at smartphone chips: the company is allegedly working on chips for Chromebooks as well, and they could arrive as soon as next year. http://https://liliputing.com/wp-content/uploads/2021/09/pixelbook-go-700x389.jpg There aren’t many details about Google’s upcoming laptop processors, aside from the fact that they’re expected to be based on ARM architecture. But moving into the laptop/Chromebook chip space would be a logical next step for the company that develops two of the world’s most widely used operating systems: Android for smartphones and Chrome OS for laptops and (a few) desktops. News from AMD market share dropping relatively more than Intel's market share is dropping. All things are relative ----- Intel has slightly better chip supply at the moment, so Apple taking share from x86 processors is actually hitting AMD harder than it is hitting Intel. AMD is also not actively playing in the M-1, the Core i3 , the Ryzen 3 market share arena, AMD is simply letting the smaller players duke it out while preserving AMD's relatively scarce chiplet supply for the market segments that count the most to AMD. What is getting reported is that AMD is losing share relative to both Intel and to NVIDIA ---- and yes, this is true enough given that TSMC is shorting AMD relative to Apple and Apple is simply taking some x86 market share in the most generic non-specific general broad stroke terms. AMD is also hording some of their most current chiplets, making sure they have enough Gen 4 chiplets at introduction time to make certain processors "pop" more when they are introduced. Otherwise, AMD will get accused of "making paper launches" when Ryzen 4 comes out in a few months. TSMC is not able to do everything like they used to, and this time around AMD is on the short end of the allocations stick. This will do some lasting harm to AMD despite AMD having the better processor designs right now. SPECIAL NOTE from three weeks later ........ TSMC quietly released several lots of AMD key chiplets, acting to moderate the shortage hit on AMD at lease temporarily. =================================================== On the positive end of things, AMD is thinking about using their limited wafer allocations to make Ryzen 4 first wave chiplets and simply skipping over the last of the Ryzen 3 very moderate advancement generations. Rolling forward to the next generation is the smarter AMD move, I think. Doing so will put Intel on their back foot again as they have no carefully scripted set of lies built up to counter that move. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/04/21 at 06:39:30 https://liliputing.com/2021/09/lilbits-microsoft-breaks-windows-11-with-an-ad-apple-is-doing-something-with-risc-v.html https://www.tomshardware.com/news/apple-looking-for-risc-v-programmers http://https://cdn.mos.cms.futurecdn.net/bHmaJcKjvkGMQVwrzzQeKd-970-80.png.we Apple is currently looking for experienced programmers with detailed knowledge of the RISC-V Instruction Set Architecture (ISA) and Arm's Neon vector ISA for its Vector and Numerics Group (VaNG) within its Core Operating Systems group. Apple's VaNG is responsible for developing and improving various embedded subsystems running on iOS, macOS, watchOS, and tvOS. "You will work in a SW and HW cross functional team which is implementing innovative RISC-V solutions and state of the art routines," the description reads [emphasis added]. "This is to support the necessary computation for such things as machine learning, vision algorithms, signal and video processing. Push the state of the art in low level computation and drive them towards energy efficient and high performance implementations by tightly integrating software and hardware." Currently, Apple has dozens of products across multiple product lines. For example, high-performance devices like MacBook laptops, iPhone smartphones, iPad tablets, and Apple TV set-top-boxes are based on custom system-on-chips (SoCs) that use highly-customized Arm cores. In addition, devices like Apple's Watch, Airpods, and Homepod Mini use system-in-packages (SiPs) powered by technologies from Arm. Apple also uses Arm cores inside its controllers (e.g., T2, W3, U1, etc.). Western Digital best indicates why companies like Apple are going to be working with RISC-V. ARM charges too much to use ARM's IP, really folks find that RISC-V being basically free to use is much much less expensive. Toss on top of that the ARM IP buyout quagmire NVIDIA is promoting at the moment and you can see why folks are abandoning ARM like crazy. With RISC-V you at least know where you are at the moment. ================================================== Apple is currently more data per cycle efficient and much more power efficient compared to either of the x86 boys. Apple's M-2 (this fall) is going to come out of the gate as Industry Leader on anything that runs off a battery. Industry Leader on battery devices BY FAR, it isn't even going to be close. M-2 has M-1 derived efficiency and all the larger compute power of the new super Apple compute cores. PCs are a different story --- they are coming across as being totally thermal limited as the core counts and power draws from Intel are simply immense and are getting bigger. AMD is headed that way too as they plan to put the AI and the graphics card distributed per chiplet all throughout their entire AMD line up, which means more heat evenly distributed across all the entire product line. Yes, AMD gets hot too, you are putting a gaming graphics card inside your CPU after all ....... AMD's new socket A-5 tops out at 170 watts. That is a big jump from 105 watts for the A-4 socket. :-/ ....... Wanna buy the new inside-the-chipset liquid based super cooler setup, anyone ???? ........ WHILE ON THE TOPIC OF STATE OF THE ART PACKAGING TECHNOLOGY ....... TSMC has built 4 brand new packaging plants in Taiwan, plants that (for example) can take your first lots of TSMC contracted chiplets and VERY QUICKLY build out the finished processors quickly while using TSMC patented feature sets like vertical via connection stacking, inside the CPU liquid cooling and full die sized memory layers and AI accelerator layers, etc. etc. etc. TSMC has a lot of new neat tech to sell you and they can stack it up for you up to 20 layers tall if needs be. This sort of activity explains how AMD can suddenly and magically make entire product line roll outs appear seemingly out of nowhere. TSMC has built this tech mainly for Apple and then for AMD, but since Intel has gone with chiplets from TSMC they can wind up using it too. Many of the very neatest 5nm AMD tricks announced in the last 6 months actually belong to TSMC. All of the now delayed 3nm advancements announced by everybody belong to TSMC as well. Ditto for Intel's much hooted and promised new 3nm tech features, they all actually belong to TSMC. When Intel starts referring to your new tech as "Intel tech" TSMC needs to really watch out for Intel's sticky fingers. Intel will flat steal your tech from you in a heart beat after they watch you use your tech to make their products for a lot or two and they finally understand how it works. Price Increases from TSMC TSMC has said they need to raise prices due to their own supply chain increases and their need to build lots of brand new facilities next year. TSMC has raised pricing for Apple (who paid up front for TSMC's original ASML machines and has paid up front for each of the newer lithography waves that have come on since then) a piddly 2-3% surcharge. AMD has been 100% on board with TSMC since Ryzen started up and AMD gets a 5% surcharge. Intel and all the rest ------ a 20% surcharge. TSMC by its surcharge rate is setting the processor dominance order for the next 5 years ........ Intel as a competitive fab is not on up there top of things, you will notice. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/11/21 at 06:11:18 Intel --- what is it good for? |
|
Title: Re: AMD & others --- post AMD dominance in 2020 Post by verslagen1 on 09/11/21 at 07:45:07 https://www.youtube.com/watch?v=bX7V6FAoTLc |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/18/21 at 16:06:16 Well, world-wide the status quo for this month did not shift for much. Intel has rolled its Rocket Lake now and did not take any market share because of it. The Rocket didn't roar, in other words, it fizzled. AMD did lose a point or two of market share simply because TSMC could not supply all the wafers of chiplets AMD wanted. Being mindful of this constraint, AMD has been building up and has been hording their next generation's wafers so as to have enough chiplets to actually do a real AMD introduction wave. AMD does not care for having any accusations of a "paper launch" similar to the reviewer blasts Intel has been getting lately on the Rocket Fizzle Lake. AMD is now 100% using TSMC's newly built assembly plants to put their chiplet based processors together as it is cheaper for AMD to do so and TSMC can blend in their patented new assembly tricks right into the AMD finished product. For example the very fast and quite massive L4 cache on top of the processor set is worth a 15% throughput improvement all by itself, and this is additive to the 30+ percent improvements coming from from lithography and architectural changes that are coming to us with this AMD generation. Yes, Rumor Sez that AMD/TSMC is going to skunk Intel totally with their new generation Ryzen processors which will begin shipping in earnest very soon ........ using TSMC as their sole source lithography and packaging partner AMD has now been given some additional wafer allocations at TSMC to allow their next generation of products a free reign to expand into new markets and uses. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/19/21 at 19:12:41 https://www.tomshardware.com/news/intel-accused-of-infringing-finfet-patent-loses-6th-challenge-against-chinese-academy-of-sciences Speaking of Intel smelling like a skunk ........ Intel Accused of Infringing on Chinese FinFET Patent, Intel Loses 6th Challenge Against Chinese Academy of Sciences By Paul Alcorn about 14 hours ago Seventh time's a charm? Intel having lost 6 times in a row in Chinese Patent Enforcement Court now, so now the Chinese is now talking about blocking all shipments of Intel's Core i processors from coming into China until the $$$ 31 million dollars $$$ in current judgments are paid in full. Intel has racked up a lot of "not-friends" over the years world wide due to their harsh legal antics ...... some of this antagonism is coming home to roost now that Intel is operating illegally in the turf of other people's patents and the other guy's national court systems. http://https://cdn.mos.cms.futurecdn.net/4DjSNkDcLstmpxNAaCgPBE-970-80.jpg.we The FinFET has served as a foundational piece of technology for Intel's processors since 2011, serving as a key ingredient in nearly every processor it sells. However, Intel has been embroiled in a patent infringement lawsuit in China since 2018 with a Chinese government-funded R&D lab that asserts the company has violated its FinFET patent. Intel has responded by challenging the validity of the patent, but it recently lost its sixth challenge with the China Patent Reexamination Board, marking yet another setback for Intel in the case as it looks to avoid a ban on the sale of its 'Core' family of processors in China. The Institute of Microelectronics of the Chinese Academy of Sciences (IMECAS) filed the lawsuit against Intel in the Beijing High Court in 2018, seeking 200 million yuan (roughly $31 million USD) in damages plus the cost of litigation. More importantly, the lawsuit also seeks a ban on the sale of Intel's 'Core' family of chips that it uses for its client products, at least until the two parties can come to a licensing agreement. Intel provided the following statement to Tom's Hardware regarding the latest developments: "This decision is subject to appeal and Intel looks forward to the courts’ balanced consideration of Intel’s invalidity challenges to the asserted patent." IMECAS has also filed two other patent infringement lawsuits against Intel (more below). Intel can't assign a dollar amount to the potential total damages, saying, "[...]we are unable to make a reasonable estimate of the potential loss or range of losses, if any, arising from these matters. We dispute IMECAS's claims and intend to vigorously defend against them." Intel's latest setback in the FinFET case follows a long string of the company's attempts to bring the Inter Partes patent review process to friendlier climes at the U.S. Patent and Trademark Office. However, the USPTO has refused to hear the case, leaving the dispute in the hands of the Chinese patent authorities. Intel began using FinFET designs with the debut of its third-gen 22nm Ivy Bridge processors back in 2011 and continues to use the tech in its latest products. The dispute centers around the 457 patent (CN 102956457), also known as the "FinFET patent." IMECAS also sued Dell China and JingDong Century Information Technology (JD) for infringing the '457 patent, but those companies tendered indemnity to Intel. Intel has encountered some success by challenging the validity of the patent — three of the claims in the patent have been deemed invalid. However, eleven of the claims are still enforceable. That means, barring any additional successful challenges, the infringement lawsuit against Intel will move forward. IMECAS has expanded its efforts, too, by suing Intel in two other patent infringement actions related to the sale and manufacturing of its Core i3 processors (CN 102386226 — '226 Patent), this time related to MOSFET technology. These actions also demand injunctions and reimbursement for litigation costs, but differ in that IMECAS reserves the right to claim an unspecified amount of damages. Intel will obviously exhaust all options as it fights the infringement lawsuits, but IMECAS has the advantage of being on its home court (quite literally). Additionally, according to ICsmart, with over 5,000 patents filed in China and 500 filed in foreign countries, along with 1,505 patent licenses in IC technologies, IMECAS has plenty of experience with patent law and is a seasoned litigant. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/23/21 at 00:55:30 https://www.european-processor-initiative.eu/?p=subscribe https://www.tomshardware.com/news/epi-deliver-epac-risc-v-processor https://www.nextplatform.com/2021/09/22/europe-step-closer-to-native-risc-v-reality/ http://https://cdn.mos.cms.futurecdn.net/PQzYCKXEAf3NTqGGLmzwCW-970-80.png.we 10 countries in the EU are funding an EU initiative specifically to make these, an EU controlled RISC-V mainframe/workstation combo chipset which will be built in 3-5 separate socketed chipsets (#2 of the set is shown) which will then mount on a multi-socket custom RISC-V motherboard (which would then be making up a readily upgradeable component pull & replace system). The series #2 RISC-V processors shown are a design containing multiple special-purpose accelerators, all centered around the RSIC-V ISA and its design principles. The series #2 processor shown contains four tiles of Vector Processing Units (VPUs) made up from Avispado RISC-V core designed by SemiDynamics, and vector processing elements design by Barcelona Supercomputing Center and the University of Zagreb. In each tile, there are home nodes and L2 cache for cache systems, which are the contributions of Chalmers and FORTH. For additional acceleration, there are Stencil and Tensor accelerators (STX) engineered by Fraunhofer IIS, ITWM, and ETH Zürich, and the variable precision processor (VRP) deigned by CEA LIST. For keeping all of these components connected, there is a high-speed Network-on-Chip (NoC) router and SERDES developed by EXTOLL. RISC-V is certainly moving and developing more quickly now. ARM is being functionally hamstrung by the NVIDIA takeover attempt and ARM and the takeover are getting stalled intentionally by various reviewing regulators (including the Chinese regulators) who keep finding reasons not to even look at the deal. Now obviously the Europeans have decided on their own future pathway, which is RISC-V based. Until their EU fab is actually built, the European consortium is shipping sample chipsets as we speak that were built by Global Foundries. AMD is doing fairly well at the moment, but if the world goes all RISC-V how will AMD keep their current technical lead? And Intel, what does it bring to a RISC-V party ????? Or, as asked earlier, Intel -- what is it really good for? Intel does not own the patents nor are they fully licensed for what Intel is currently building and Intel is now getting hammered by the current waves of Chinese litigation accordingly. For details, look at the post above this one, but what sane customer would want to hop into this legal meat grinder with Intel knowing Intel is not able to protect them? I do not think Intel will get much traction as a foundry once they screw over a few orders by not building what the customer wanted when they wanted it --- but instead using some weird Intel custom methodology resulting in it not working as intended ..... or by simply failing to build it on time. Intel always gives priority to their own production needs over any fab customers needs which always results in "failing to build it on time". These are Intel's favorite failure modes from the last 3 times they tried to be a fab ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/25/21 at 02:07:46 https://azgovernor.gov/governor/news/2021/09/intel-breaks-ground-two-new-semiconductor-factories Intel breaks ground in Chandler Arizona to build 2 new fabs. Intel's competitors TSMC and Samsung are building complete at this time and are putting in their equipment to crank up sometime in early 2022. Intel won't begin production until 2024. At this Intel level of "don't get it done" the new Intel plants may never actually come to pass. They will be dropped due to some ostensible good reason before Intel even finishes spending the Arizona seed money. Intel's last plant (Fab 4) was delayed like this over 10 years before it was completed. Right now the chip shortage will bottom out and the entire chip market will roll towards an oversupply situation at least one full year before the Intel plants are scheduled to open. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 09/29/21 at 09:50:26 https://www.tomshardware.com/news/amd-increase-efficiency-of-chips-thirtyfold-by-2025 https://www.phoronix.com/scan.php?page=news_item&px=AMD-30x-Energy-Efficiency-2025 https://www.pcmag.com/news/amd-aims-to-make-supercomputer-chips-run-30-times-more-efficiently-by-2025 Increases like this can be expressed as battery life increases or as in speed increases (providing that appropriate cooling is available). Or a mixture of both if that is what is needed. Please remember, AMD now has access to a whole new world of Xylinx accelerators and FPGA technologies, along with a several new matching sets of TSMC processor and packaging technologies. AMD having formally announced this 30 TIMES IMPROVEMENT to completely fall due inside 4 years from now it becomes easy to predict a long series of smaller improvement steps that will keep AMD on top of the computer game. This also indicates that AMD has a future pathway planned out that exceeds Intel's most fervent wet dreams. AMD also habitually hits both their goal numbers and the production due dates, both with great regularity ---- something that Intel fails at more often than not. ;) I can vaguely see a future where AMD is no longer x86 dependent, with AMD announcing new standards groups and new cross license agreements with Samsung and Mediatek and others, intentionally ending their dependencies on IBM and Intel within 4-6 years. Let's be honest, both ARM and RISC-V are far far far more energy efficient than x86 could ever hope to be. And these are the least efficient of the new technologies that are coming out now. =================================================== https://www.tomshardware.com/news/amd-we-stand-ready-to-make-arm-chips https://www.amd.com/en/amd-opteron-a1100 https://seekingalpha.com/news/3740563-amd-shares-climb-as-company-eyes-making-arm-based-chips Long story short, both AMD and Intel are frantically paddling to create ARM based and RISC-V based server chipsets. Why? That huge 30x improvement in calculations per watt has become a priority with both of the old x86 companies. Both realize x86 cannot compete against ARM and RISC-V based server chipsets. AMD is working with Samsung and with ARM and is a founding member of the RISC-V based server chipset consortium. More news that is new ---- Microsoft and Mediatek are focusing on AMD as a key partner for the next generation of progress. Why AMD ????? Because Intel is clearly broken and it cannot innovate enough to stay ahead of AMD or ARM right now. AMD and ARM get along quite well, while Microsoft recognizes a need for a Consortium to use ARM processors in servers simply to get the lower energy costs and greater efficiency that ARM provides. AMD as run by Lisa Su is trusted to keep their word in a business relationship. Lisa Su can bring a lot to the table in a consortium arrangement, and I do mean a lot ----- Lisa brings really strong AMD graphics and now also has all the Xylinx accelerators and AI functions that Xylinx invented and owns the patents for. All of the real modern 64 bit x86 patents belong to AMD and AMD shares this tech base with TSMC who now can do up a chiplet based processor systems package using AMD tech and their own chip packaging refinements which are now quite considerable. In contrast, note that TSMC has started shunning a closer relationship with Intel due to Intel's failing to keep confidences TSMC shared with them (indeed Intel has calling out TSMC's tech as if it were their Intel tech, something that no oriental mentality would care for). Intel thinks because Joe Biden needs Intel, that makes Intel powerful worldwide. Joe Biden gets laughed at a lot worldwide, and Intel gets their own sets of chuckles as well for being so close to Biden. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/04/21 at 22:07:55 https://liliputing.com/2021/10/windows-11-is-now-available-on-some-devices.html Time to line up to get your pocket picked again, sez MickeySoft. This time Windows looks like Apple ....... but it is still slow and buggy and bloated like Windows always is, but under a pretty Apple looking shell. To get it all to work (and to get it to work right) you will have to buy a new very expensive computer. ::) When Microsoft first launched Windows 10 in 2015 the company said it would be the “last version of Windows,” since the goal was to continually push out incremental updates (more bloat) rather than revolutionary new version of the operating system. Today Microsoft has released Windows 11 for "security reasons". There’s no word on whether this will be the last version of Windows, but the update does bring significant changes to Start Menu, Taskbar, Microsoft Store, and a number of other areas. So I guess it’s not surprising that Microsoft decided it was time for a new name… (and a new payout from your wallet to MickeySoft's coffers) although there’s some debate over whether we really needed some of the new things Microsoft is bringing to Windows. http://https://liliputing.com/wp-content/uploads/2021/10/win11-768x432.jpg Among other things, Windows 11 brings: There’s a new Start Menu with access to apps, search, files, and other data. It’s centered by default, but can be moved to the side of the screen. Windows Taskbar now features icons in the center rather than the left side. When using a tablet without a keyboard, the UI will also adjust with larger touch targets and more space between icons. Widgets are back, but this time they’re not called Gadgets and they’re not on the desktop, but rather in a card-style feed that you can access by swiping from the edge of the screen. Snap Layouts give you more ways to organize application windows on your screen. And Snap Groups let you maximize and minimize a whole set of apps together. Microsoft Teams Chat is integrated in the operating system, whether you want it or not. The new Microsoft Store has a streamlined design and now supports Win32 apps as well as Universal Windows Platform apps and Progressive Web Apps. There’s even support for third-party web browsers in the store now, and for third-party app stores like the Epic Games Store and Amazon Appstore. Microsoft is bringing support for Android apps to Windows 11, but it won’t be available on day one. Eventually you’ll be able to download Android apps and games from the Amazon Appstore or sideload them, and run them alongside native Windows apps. Auto HDR bring high dynamic range even to games that don’t natively support it when you’re using an HDR compatible display. DirectStorage brings faster load times to games on systems with NVme storage and DirectX 12 graphics cards. To fully understand Win 11 you can let our old buddy ArsTechnica help you do a really deep dive into Win 11. https://arstechnica.com/gadgets/2021/08/why-windows-11-has-such-strict-hardware-requirements-according-to-microsoft/ https://arstechnica.com/gadgets/2021/10/windows-11-the-ars-technica-review/ |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/06/21 at 07:22:22 https://liliputing.com/2021/10/intel-alder-lake-p-details-leaked-mobile-processor-with-14-cores-20-threads.html https://wccftech.com/intel-12th-gen-alder-lake-mobility-cpu-roadmap-confirms-alder-lake-p-with-14-cores-alder-lake-m-with-10-cores-first-laptops-with-ddr5-support/ http://https://cdn.wccftech.com/wp-content/uploads/2021/08/Intel-Architecture-Day-2021_Pressdeck_Final_EMBARGO-compressed-069.jpg Alder Lake is a huge chipset, likely still being run at 14nm ---- and please note the initial releases of Alder Lake did not outperform the old ATOM versions of the mobile chipset that Intel already had out at the time. This will likely be somewhat better, but not commandingly so. Move along, nothing to see here ......... :o WOW !!!! Win 11 sucks so bad it overcomes any benefits from the new Intel processor, throwing Wintel as a team back 50 yards or so for a clear loss. I stand corrected ---- a 15-25% decline in performance and a large INCREASE in security hits is really not "nothing to see here". Windows 11 is racking up reviewer condemnations by the dozens right now and some of the performance hits they have found are significant, especially in gaming. https://www.google.com/search?q=windows+11+loses+15%25+performance&oq=windows+11+loses+15%25+performance&aqs=chrome..69i57.22667j0j4&sourceid=chrome&ie=UTF-8 Just pick a few at random and read up a bit ...... Wintel is way way WAY back on their back foot now, promising all sorts of "fixes" and "patches" out the wazoo when faced by the reviewers ire. AMD shares this pain on their notably better running processors, BTW, just slightly less severely as AMD had no direct hand in causing this Wintel disaster. ...... did they simply not beta test this Win 11 stuff before releasing it ????? And for this large pile of brown smelly bullshite the Wintel boys coerced you into buying a brand new expensive PC ????? |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/13/21 at 11:47:40 We have us an historical moment in the past 3 years that is occurring right now ----- Intel has three chipsets out of their entire current crop of chipsets that can claim "competitiveness" against the currently released AMD products ...... if you hold your mouth just right, that is ...... and compare them to a slightly non-standard match up that is giving the advantage to Intel. Intel does have some issues that MS just rang the combined Wintel effort 15-25% back down into the hole with a very strongly flawed Win 11 initial release but hey, when you are Intel you have to take whatever crumbs you can get when you can get them. This is a lot closer than Intel has been in 3 years ...... But sorry, Intel, there are no awards for close second place. Hey Lookie Over There, Intel --- AMD has got two entirely new generations of their entire AMD line up sitting over in the wings waiting to ship out very early first quarter 2022 and then again later third quarter next year (processors and packages in the first wave have already been produced in TSMC's new facilities no less). They both swing HUGE 1.5 gig caches on the top of the processor stack, yep some right huge brand new TSMC tech 15% product improvement L3 caches ..... and these chip assemblies are also carrying some 25-35% throughput TSMC architectural improvements inside the processor itself all across the board. ...... AND A BRAND NEW TSMC LITHOGRAPHY SHRINK OR TWO are scheduled for 2022, let's not forget those ....... Intel will go back in the hole rather deeply yet again as each new AMD wave is released and the products go out for benchmarking. And TSMC has also come up with some additional AMD 6nm and 5nm in the first quarter of 2022 and some additional 4nm and 3nm production volume allocations for AMD to use in the 3rd quarter of 2022, so AMD can actually go peel off some more fresh Intel Market-share each month in 2022. ::) Fair and balanced coverage Intel has stronger commercial agreements with their box makers, to say the box makers trust Intel to be the better option in the long term is an industry truism. Intel can screw up royally in the short term and the box makers just keep on using their current stuff on the theory that Intel will fix it next time. They also trust Intel to spread enough brown vapor around such that consumers will buy the stuff when the box makers deliver it. Intel does ink some very restrictive distributor agreements that are in restraint of free trade and they are allowed to get away with it by US regulators. EU regulators fine Intel every year or so and every year Intel stalls paying the penalties, endlessly stringing the legal process along until the judgments become moribund. The box makers will always support Intel BECAUSE THEY REALLY HAVE NO CHOICE BUT TO DO SO. Watch the budding RISC-V roll out in the EU break up a few of these rigged games, but also watch Intel and MS take legal actions to protect their duopoly in the long term. Dirty Tricks Department The new AMD L-3 cache (based on TSMC packaging tech) isn't recognized by Win 11. Why? Because it wasn't invented by Wintel and Wintel has no CURRENT answer to the 15% power boost it gives to AMD processors. So, jest go break Windows so the new AMD stuff doesn't work right, that seems to be the answer the Wintel boys always seem to have for innovations coming from somebody else. Please note that older versions of Win 10 still works fine (this week anyway) and Linux's increased performance is unchanged as well. UPDATE: It turns out that Microsoft had already been given the correct drivers by AMD, but MS simply hadn't rolled them out in time for the Win 11 introduction. oops :-[ MS has caught a whole bunch of shite over this "unintentional omission" and has corrected it as of this week's software roll out. Intel can no longer claim any form of equivalent performance as the AMD 1.5 gig L-3 cache yields a clean 15% performance boost for AMD's current production processors. Intel got to see the bright sunlight for a day or so on 3 processors, but now it it's getting stuffed back down underneath the big flat rock again --- poor old Intel ...... =================================================== Intel has finally broken ground on their two new fabs in their Chandler Arizona complex this past week. Samsung and TSMC have their buildings up already and are busy installing production lines. ASML is building extreme EUV machines at the rate of one a week to put into all these new plants. ASML is the working standard for all this stuff (5nm and 3nm are fully developed standards that all computers will be built to). Intel will be late on getting equipment as they didn't order it back in a timely fashion like Samsung and TSMC did. This is good for the industry as a whole as Intel vs Apple vs ARM have shown us that grossly un-mixable standards really don't help things to move along together harmoniously. If Intel is going to be a fab they need to GET ALONG with everybody and work within the current most modern standards instead of always "doing it the Intel way ......." and going off on their own as they always did before isn't going to work any more. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/19/21 at 20:23:22 https://liliputing.com/2021/10/lilbits-raspberry-pi-legos-alibaba-open-sources-its-risc-v-chips-mac-pro-could-have-40-core-cpu-and-128-core-graphics.html Alibaba open sources its RISC-V chips, Mac Pro could have 40-core CPU next year and 128-core graphics too http://https://liliputing.com/wp-content/uploads/2021/10/710-700x500.jpg This is a big deal as it shows both RISC-V and ARM servers picking up real muscle power to take on the x86 boys and to simply slay them with power efficiency and superior processing power. Alibaba Cloud (China) Unveils New Server Chips to Optimize Cloud Computing Services Alibaba introduces Yitian 710 server chips with up to 128 Armv9 CPU cores. The company also plans to open source the code for its XuanTie RISC-V processors cores as well a Linux, Android, RTOS, and AliOS software stacks for those chips. It looks like source code and documentation for E902, C906, and C910 are already on GitHub. And being open source on BOTH hardware and software, anybody can use it ....... Intel and AMD both took a lick on this one. Open Source RISC-V processors are quickly becoming more competitive to ARM mobile chipsets and both are edging closer to being competitive to Intel/AMD. Alibaba and Google and Amazon Web Services are showing up now as major customers/competitors for mainframe processor designs. I will communicate the closing race between the processor types as they each develop down their relative pathways. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/22/21 at 17:55:06 https://liliputing.com/2021/10/sifive-says-its-next-risc-v-processor-outperforms-arm-cortex-a78.html When RISC-V chip designer SiFive introduced its Performance P550 processor in June, the company said it was the highest-performance processor based on RISC-V architecture to date. Now the company is previewing a new processor that SiFive says is 50% faster than their previous efforts, allowing the new chip to theoretically outperform an ARM Cortex-A78 processor. http://https://liliputing.com/wp-content/uploads/2021/10/sifive-preview_01-768x432.jpg SiFive’s announcement is a bit light on details at the moment, the company hasn’t even given the new chip a name yet. But the company says more details will be revealed at the RISC-V Summit in December. So far we know that SiFive’s next chip: * Supports up to 16 CPU cores * Runs at frequencies up to 3.5 GHz * Features 16MB of L3 cache, up to 2MB L2, and * SiFive says the new chips are 64-bit processors with quad-issue out of order processing, which can be scaled to support a wide range of devices including PCs, servers, or mobile and embedded devices. SiFive says the new chips are 64-bit processors with quad-issue out of order processing, which can be scaled to support a wide range of devices including PCs, servers, or mobile and embedded devices. High-performance versions would likely combine a cluster of chips together, offering up to 128 CPU cores. This chip and some other planned new ones was why Intel was so hot to buy up SiFive a few months back. SiFive has backed away from the deal totally now and no longer plans to use Intel as their exclusive one and only foundry as Intel's attitude and recent actions had left a bad taste in their mouth. Intel has a whole lot to learn about being a fab and about remaining "properly neutral" about it ........ =================================================== one week later Intel has posted their stock holder guidance for the next quarter and Intel IMMEDIATELY lost 10% of their stock value in ONE SINGLE DAY as Wall Street downgraded Intel according to their own guidance information. Why? Reading between the lines on the attached fiscal information it was clear that Intel didn't have the ready funds needed to execute all the various things they had already committed to do early this upcoming year. It also became clear why Intel chose to drop their SiFive buyout bid and their Global Foundry purchase bid so abruptly. All of Intel's pin money has to go towards the equipment to put in the two new fabs in Arizona. :( Intel isn't broke yet by any means, but Intel lacks the sufficient very very large positive cash flow "free money" they used to have in years past. Intel is also getting downgraded by the futures traders because relative to their current COMPETITORS (and they have a lot more competitors than just AMD) the performance of what Intel builds now at 14nm is beginning to smell more like sardines left out on an aluminum boat seat in the hot sunshine. Yep, it stinks like very very dead bait fish ........ Intel has no winners out there at the moment. Intel has no credible plan to have a up and coming winner in the short term either. Intel has no plans for any long term planned winners laid out in any credible fashion going out for next year or so (apart for getting TSMC to build some selected 3nm stuff for them, actually). TSMC building some stuff for Intel at TSMC 3nm is Intel's next best window to achieve anything close to a "winner" product. Intel will have to continue to lower their 14nm general processor prices even more to be able to move their "non-preferred" 14nm product line going out into the foreseeable future. So, Intel is facing a short term "crisis of confidence" both on Wall Street and in the actual market place. This will all get worse over time until Intel starts building something at a modern competitive lithography level and comes up with some winning new designs. :-/ (poor Intel, so sad) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/25/21 at 05:13:15 https://liliputing.com/2021/10/intel-launches-12th-gen-alder-lake-chips-for-desktops-up-to-16-cores-24-threads.html Intel has begun their next brown vapor campaign, this one is based on using their Big, followed by their "middle" and then the smaller cores set ups. Yep, up to 16 cores per unit being tested is what you are seeing the results for. http://https://static.techspot.com/images2/news/bigimage/2021/10/2021-10-26-image-37-j_1100.we Intel has a software that puts the loads on the currently coolest Big processor core, switching them out as they heat up. This is a neat trick, and when it finally works right it should help out with better real world throughput results ....... Please realize any intentionally leaked results you see now a days from Intel are based on the best core being measured for the single core results and then being run out to the same size core count of the entire chipset to make up the many proposed many core results. This is not how it actually works in the real world ........ Intel marketing is systemically lying and cheating again. First, only one core out of the entire set actually does work that good, not 4, not 6, not 8 and certainly not 16. The little cores are not able to do the work of the middle sized cores much less the amount of work that can be done by the big cores. We are now seeing that the smaller cores are actually all the same thing, simply ranked and split up by the Intel processor management software, they were originally intended to all be middle cores, but some didn't make the grade and the Intel processor management software keeps them sorted. Forget about running all of this at once in a mobile product. 300 watt combined power draws running the cores all at once is NOT A REALISTIC SCENARIO on any mobile product. The new LGA1700 socket to hold this 10nm monstrosity is simply HUGE for a mobile processor equipped product. It is bigger than the old Intel desktop socket used to be. Intel sees this as a plus for cooling, since their new processor management software shuttle bumps the processing load around all over the large chipset to keep things as cool as possible. For both the new Intel and the new AMD products, look to your cooling methods as all of these new chipsets are cooling thermal limit bound at the top end of their performance range, even if they could theoretically churn out these huge potebtuak outputs with no regard to their internal thermal limits. I would say we are getting into the realm where inside the chip liquid cooling is going to be needed by both Intel and AMD, so look for results which are made using real processors in real laptops using a fully identified stock cooling system (whatever it is). (use of freon based cooling systems by some test houses has been both suspected and has been occasionally verified by catching them red handed, yes, Intel we are talking about you at the big sales show a few years ago. Intel had a freon rig the size of a small dorm refrigerator tucked away under their display table and their demo shown on their display screen was actually being run on a dual core mainframe motherboard with a dual set of Freon chilled mainframe processors ...... ) :P Next, place no confidence in the results of any small custom test houses. One reviewer called them "hired liars" and he wasn't far wrong in that estimation. Lastly, you don't really need the most gee whiz best thing out there ---- Linux actually runs faster on lesser processor sets, as does older versions of Windows. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 10/29/21 at 07:30:54 Folks, due to a hard hitting Covid booster shot my wife and I are ailing a bit. Not dying, but not feeling well. Expect fewer comments from me for a while ....... AMD just posted some very good market share and financial return numbers. Intel did not, and Intel has lost 11-12% of their total stock value per share recently as a result. Pundits are reviewing the Intel presentation of Alder Lake and the public is asking questions like "Is it safe to swim in Alder Lake at this point?" So far their overall answer is " NO '" as Alder Lake is actually pretty green and scummy. Seems appropriate that the lake and the processor both share a similar review, huh? Intel is suffering strongly from their own Marketing BS and AMD is busy still taking more Intel market share as TSMC has found quite a few extra wafers to supply AMD's market share takeover efforts. Intel also released fiscal information that leads some to think they are getting cash strapped by the two Arizona plants they are building ----- some are even saying internal Intel documents refer to intentionally pausing building up the new lines inside the buildings until Intel demand picks up enough to keep the new lines 100% busy. Joe Biden had promised Intel a lot of USA gov money that is not readily apparent at this time. Instead the USA gov is asking Intel for contractual default money payments off the Frontier supercomputer project, the one from two years ago that Intel totally dropped the ball on. Intel's plans to build a German or Italian plant are now seen as quite dubious accordingly ....... All of these plans are contingent on the country in question providing massive tax breaks and contributions. TSMC, Samsung and Intel all require lots of up front money from the state were the stuff is going to be built in. They aren't getting it right now from USA gov as the Democrats have spent it all elsewhere on stupid stuff. Expect all these Fab guys to hit pause until the funds flow again ..... ================================================== BIG PICTURE TIME Intel has created their own crisis recently by renaming everything, suddenly saying that their old 10nm is now 7nm and 4nm and by "creating" lithography modes that have dubious benefits. This creates distrust and confusion with chipset buyers. Wall Street is crushing Intel for their flaws, taking away stock value by the week. Meanwhile, AMD and TSMC are making real progress by carefully following their published road maps. Intel is having to use lots of brown vapor BS to make up for their lack of progress over the last 3 years, creating new roadmaps to disguise the lack of completing the old ones. Intel is likely going to bankrupt these split off company divisions, ie. have to stop building on the internals of their new plants due to being late to the party and NOT HAVING ANY ORDERS TO RUN IN THOSE PLANTS TO KEEP THE PLACES BUSY ENOUGH TO PAY OFF THE LOANS. Both TSMC and Samsung are deeply committed to building out what they have started, but you can understand their reluctance to keep putting their money into anything if the USA gov money (local and federal funds) do not flow as promised. Next opportunity for Intel to create some form of "real winner" comes in 1 year from now when Intel has TSMC to build a line of processors for them at one of TSMC's 3mn advanced nodes (which Intel will call something else that is complete BS of course). Cooling is now critical for everybody as all state of the art processors are growing in power and complexity. 300-600 watts to drive a chipset means a significant chipset cooling system is needed as well. I am losing all interest as I cannot see me ever buying any of this state of the art stuff as the cost of it has gone all vertical on me and my computer discretionary funds are getting slimmer and slimmer. Actually, my real computer needs are getting less and less as well ....... If you are buying used, buy a cheap used Dell workstation machine as they are all well understood and they have all the software bits and pieces available in a stock Linux OS system. If you are buying Windows OS machines from Wintel, good luck and Godspeed to you, brave pilgrim, as the Wintel boys are forcing you to buy a very expensive brand new machine to meet their new security requirements ...... Both Intel and AMD are going to chiplet based Big Little designs with larger core counts than you are used to seeing. There will be some AI boosters stuck in the heterogeneous core mix and on-board graphics good enough for a worker bee's needs will be built into all of them. All of these will be quite expensive and far far more powerful than anything I need. ::) The world economy is being crippled right now due to a lack of shipping throughput when the container ships hit ports and cannot be unloaded in a timely fashion. Covid is being blamed for all of this but Covid was not all of it as other root causes and a general recession are playing out as well. Governments will eventually lack the money printing capacity due to dilution and will not be able to support all of this with more and more large gov checks to people who don't want to work. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/02/21 at 12:24:37 Welcome to the Angstrom Era PS: Moore’s Law is Still Dead by Steven Leibson Intel took a lot of flak when it announced new names for its future semiconductor process nodes during the Intel Accelerated event in Late July. The new nodes are called Intel 7, 4, 3, and 20A. Industry pundits have knocked the company for calling its 10nm enhanced SuperFin process node “Intel 7.” (Intel announced and demonstrated 12th Generation i5, i7, and i9 Core processors built using the Intel 7 process node just last week at its Intel InnovatiON developer event, so this process technology is clearly well in hand.) The company now uses the name “Intel 4” for the node previously called 7nm. Intel 3 and Intel 20A are all-new node names. Part of this node renaming has roots in marketing, and part is simply facing reality. In the past, Intel had said that its 10nm enhanced SuperFin node (now renamed Intel 7) was roughly equivalent in power and performance to TSMC’s 7nm node. Confusing, no? The company has also stated that industry analysts have asked Intel to update its process node nomenclature to reflect the company’s true competitive position in the semiconductor process arena. All of that is the marketing part of the renamed nodes. Years ago, process node naming was based on a transistor’s smallest feature size. That naming convention was specifically developed for a planar MOS transistor, and the smallest transistor feature was invariably the transistor’s gate length. That naming convention went out the window when FinFETs took over from planar MOS transistors, and all vendors’ process node names became nothing more than estimated equivalents for the power and speed you’d get from an equivalent planar MOS transistor. Except that they couldn’t actually make those planar MOS transistors any more. Planar transistor technology gave out. MOS transistors just don’t work well at current lithographic levels. That’s why we turned to FinFETs. Intel introduced its first FinFET process at the 22nm node – way back in 2011. RibbonFETs and angstroms Ten years later, the fundamental structure of transistors is about to change yet again. FinFET gates are driven from three sides. That’s two more sides than are driven in planar MOS transistors, and which results in much better transistor performance at the expense of more complex manufacturing techniques. However, driving three sides of the FinFET transistor’s gate no longer achieves the desired speeds and leakage currents. We must now drive all four sides of the gate. These 4-sided gate-drive structures are universally called “gate all around” or GAA. Universally, except for Intel, which calls its GAA transistors “RibbonFETs,” which are currently scheduled to appear in the Intel 20A process that’s due to appear in the first half of 2024 (barring delays). Notice the “A” in the process name. “A” is for “angstrom.” Instead of calling this process node “Intel 2” as with the previous process nodes, Intel switched units from nanometers to angstroms. An angstrom is 10-10 m, or one tenth of a nanometer. I think we’re supposed to simply disregard the fact that Intel dropped the “nm” from the bigger process nodes. Presumably, this nomenclature change to angstroms allows for fractional nanometer node naming. That’s similar to the situation we had with the previously used unit, microns. Back in the dark ages, we had 3, 2.5, 2, 1.5, 1.3, 1.2, and then 1 micron process steps, followed by 0.8, 0.75, 0.7, 0.5, and all the way down to around 0.25 microns. Somewhere around that point, we jumped from microns to nanometers. Based on my best recollection, that nomenclature change occurred at 0.18 microns, which was popularly called 180 nm. When was that? Around 1998. More than two decades ago. That means the nanometer process node ruler will have been in use for almost a quarter of a century when the angstrom era starts. The Intel angstrom nomenclature allows the company to have process nodes named Intel 18A, Intel 17A, Intel 16A, and so on. That sounds so much better than Intel 1.8, Intel 1.7, Intel 1.6, and so on, doesn’t it? Sounds like they’re making more progress this way, right? However, the progress from one node to the next isn’t nearly as big as it once was. From the image above, you can see performance/watt improvements of between 10 and 20 percent from one Intel process node to the next, and the image doesn’t even discuss density improvements. Moore’s Law lies a moldering in the grave And that gets us to the meat of the matter for this article. Based on all that’s written above, I have very sad news for you. Moore’s Law is dead, despite Intel CEO Pat Gelsinger’s bold and enthusiastic promise to meet or beat Moore’s Law every year for the next 10 years, made last week during the Intel InnovatiON event. Scaling a process node by shaving a few angstroms from one generation to the next does not achieve a doubling of transistor density, and that is the true essence of Moore’s Law. Moore’s Law did not refer to transistor power or speed. Moore’s Law says that the number of transistors on a chip doubles approximately every two years. That’s all it says. You can look it up if you like. Gordon Moore’s original article in Electronics magazine from April 19, 1965 was titled “Cramming more components onto integrated circuits.” It’s conveniently posted on Intel’s Web site for you to read. This article appeared three years before Moore and Bob Noyce founded Intel, while Moore and Noyce were both still working for Fairchild Semiconductor. Moore performed an unbelievable feat of prognostication in his groundbreaking article. The semiconductor industry eventually embraced and transformed his prediction into a self-fulfilling prophecy, which was based on a very few data points. The first data point was a single transistor. The second data point was one of the first commercial integrated circuits, a 3-input NOR gate called a Fairchild [ch956]Logic Type G RTL chip. According to Dr. David Patterson, Moore’s Law lasted through 2015 when it ran out of gas. (See “Fifty (or Sixty) Years of Processor Development…for This?”) How could Patterson say that Moore’s Law died in 2015 when Intel’s executives would uniformly have you believe that Moore’s Law is alive and well today? It’s because of a phrase on page 2 of Moore’s 1965 paper. That phrase is: “…the production of larger and larger circuit functions on a single semiconductor substrate.” (Emphasis is mine.) Moore’s Law is about monolithic integrated circuits, and that’s just not where the semiconductor industry is heading. In fact, on the 50th anniversary of the International Solid State Circuits Conference, in 2003, during a presentation titled “No Exponential is Forever,” Gordon Moore himself stated categorically that “…no physical quantity can continue to change exponentially forever.” Moore’s Law was already dying nearly two decades ago, and Moore saw it himself. My current poster child for signaling the end of Moore’s Law is Intel’s own Ponte Vecchio GPU. Intel is assembling this integrated device using 47 active “tiles” – Intel’s name for the chiplets or die in a multi-chip package – manufactured by multiple semiconductor vendors from five different semiconductor process nodes, all combined in one package using 2.5D and 3D assembly techniques to produce an integrated product with more than 100 billion transistors. (Please be sure to read that last sentence with a Carl Sagan accent and emphasis on “billion.”) Some people have claimed that Moore foresaw multi-chip packaging in his article. They cite this phrase: “It may prove to be more economical to build large systems out of smaller functions…” But they seem to omit the second half of this sentence: “…which are separately packaged and interconnected.” Here, Moore was clearly discussing the use of multiple individually packaged chips on one board, a staple of board-level design since integrated circuits first appeared in the 1960s. From my perspective, Moore was clearly not predicting today’s multichip packaging with this sentence. In fact, his article discusses the possibility that monolithic ICs with 65,000 components per IC could appear in 10 years, by 1975, which was far more discrete components than any single printed circuit board could accommodate back in 1965 when Moore’s article appeared. Who could possibly need more than 65,000 components? If Moore thought about it at all back then, and he probably did, he must have seen that multichip packaging technology would not be needed until far, far in the future. Well, that future has arrived. Intel’s Ponte Vecchio GPU incorporates 47 tiles from five different process nodes to cram 100 billion transistors into one package. Multichip packaging only makes sense only because different process nodes deliver different cost/performance/capabilities tradeoffs, because we’re at the reticle limits of current chip-making equipment, and because 2.5D and 3D packaging techniques are now sufficiently practical and economical to make this approach work commercially. Why shouldn’t blocks and subsystems be made from the most efficient semiconductor process node possible, assuming you have the manufacturing processes needed to assemble all of these tiles or chiplets reliably and economically? Ponte Vecchio is admittedly an engineering marvel, but it’s most definitely not a monolithic chip, and so it’s not an example of the original Moore’s Law in action. Except for the massively mythological underpinnings of Moore’s Law, it’s really not important to most of us how Intel crams 100 billion transistors into the Ponte Vecchio package. It’s not important to systems engineers using Ponte Vecchio GPUs in their designs. It’s not important to people using graphics software or computer games that run on Ponte Vecchio GPUs. The device’s performance, power, and price (the three fundamental “P”s of all engineering design) are what’s important to those of us who live outside of the package. Intel plans to build Sapphire Rapids, the code name for the company’s next-generation Xeon processor, in a similar manner. Intel will manufacture Sapphire Rapids using four CPU tiles and 2.5D assembly based on the company’s EMIB (Embedded Multi-die Interconnect Bridge) technology. Versions of the Sapphire Rapids processor will also incorporate multiple HBM2 (High Bandwidth Memory 2) DRAM stacks in the same package. Multi-chip packaging is not unique to Intel. Far from it. AMD, Nvidia, and Xilinx all manufacture integrated devices that look like monolithic integrated circuits on the outside but are multichip devices – interconnected collections of tiles or chiplets – on the inside. (Actually, these fabless semiconductor vendors all have their devices manufactured and packaged for them by 3rd-party foundries – often that’s TSMC.) This has been the situation for more than a decade. For example, Xilinx introduced the Virtex-7 2000T FPGA in 2011. It’s based on a multi-chip package that puts four FPGA dies on top of one silicon interposer. TSMC fabricates that device for Xilinx. Around the same time, multi-chip packaging allowed Xilinx to incorporate 28 Gbps transceivers into its Virtex-7 580T FPGA before it was possible to build these transceivers directly into the CMOS FPGA die. Xilinx has expanded its use of multi-chip packaging with each new FPGA generation. That’s a good indication that multi-chip packaging works well, at least for the high end of the IC market. So, welcome to the angstrom era. It’s a “Beyond Moore” era. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/04/21 at 19:46:02 https://liliputing.com/2021/11/alder-lake-review-roundup-intel-retakes-the-crown-from-amd-for-now.html This review roll up contains all 17 benchmarks by the best in the industry, so click on it and read up if you are interested. http://https://liliputing.com/wp-content/uploads/2021/11/alder-lake-s-768x432.jpg Alder Lake had a messy introduction. It was all mashed in with the more than slightly screwed up Windows 11 roll out ........ and was also involved in AMD not getting the correct drivers issued by MS with the initial Windows 11 release. These have been corrected by MS in Win 11 nightly updates and retests have resulted in the following roll up of results which is a best a very slight win for Intel contingent on the testers using ONLY THE VERY BEST AND MOST MODERN LIQUID COOLING SYSTEMS. http://https://cdn.arstechnica.net/wp-content/uploads/2021/11/Intel-Alder-Lake-CPU-tests.005-1440x1080.png Alder Lake indeed sucks in massive amounts of power and makes for massive levels of heat at new levels that had never been seen before ....... this resulted in the COOLING SOLUTIONS used on the first released Intel Alder Lake products that simply were not adequate to allow the first samples to even equal the performance of existing AMD processors using stock AMD fin and fan coolers. BUT (and it is a big but) if you give Intel a huge power supply and an even bigger liquid based cooling system Intel does indeed eek out an slight edge over win over AMD --- temporarily --- as Intel runs through their limited thermal headroom and BIOS slowdown then occurs. Give both companies the same level of cooling, AMD always wins. AMD has two new generations of processors already sampled that will outshine Alder Lake by double digits and then later on next year by triple digits (assuming both get the same level of cooling). Going on out into the future ...... TSMC and AMD have cooperated on new packaging advances for future generations that actually have inside the chipset liquid cooling, a trick that will open up processor performance to levels never even dreamed of before. Intel will be there, too, nicking at whatever lead AMD achieves with products that do almost as well. Tit for tat, AMD and Intel will crab walk on up the performance stairway to heaven while engaging in an endless back and forth sword fight ........ FOR THOSE WHO PREFER TO WATCH RATHER THAN READ, THIS IS A GOOD VIDEO FOR YOU TO WATCH THAT ROLLS ALL OF THIS UP FAIRLY TIGHTLY. https://www.youtube.com/watch?v=dE9N95uSqHA And yes, he is talking about 36 CORES and 72 threads inside an AMD mobile processor which is potentially possible using those new TSMC lithographies and packaging techniques. When you drop down to 4nm and less, there is a whole lot of room on that AM5 socket set to put lots of chiplets (at 8 processors per chiplet no less). |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/07/21 at 22:33:06 We now have 3 new Intel Alder Lake processors announced. A pattern is showing clearly now, Intel intends to beat selected AMD processors and furthermore it intends to sell these AMD beaters at 25% less than AMD's current pricing. Intel can do this. For a while, anyway. Intel can spec out a chipset that is physically twice as large, pulls up to twice the power draw (but not for long as it overheats very badly) and even then it only marginally outperforms the AMD unit until the system BIOS shuts the processor down due to overheating. Cooling is King now for both sides ...... Intel will be happy with a win right now, no matter how marginal or energy expensive or short term in duration it might be. It is clear Intel intends to refine these competition points, expand on them and try to take some market share back from AMD no matter what it costs Intel to do this in the short term. Intel knows it has to do this now, quickly, before the next generations of AMD lithography and structure progress roll out to smash them flat again ...... Wall Street is simply killing Intel's stock value right now because Intel is not making any progress so Intel will now try to brute force some progress at any amazingly high cost levels, then Intel will have to sell the resulting new processors for less money than the AMD competition for long enough to cement their win in the eyes of Wall Street. Intel is faced with multiple "house of cards" fall down potentials right now and Intel is doing whatever they can to keep all the cards upright and still standing ....... |
|
Title: Re: AMD & others --- post AMD dominance in 2020 Post by Oldfeller--FSO on 11/08/21 at 09:58:24 https://youtu.be/ECHhuvuiNzs Lisa Su presents the next 2 years of AMD progress, with next year being only 4 weeks from now. Now you see why Intel is forcing their hand so early and abruptly, just to get it counted before it gets overcome by AMD events. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/09/21 at 03:18:57 grey bar means discontinued http://https://cdn.mos.cms.futurecdn.net/cokZwAZcmKDRcEmCyrYRnF-970-80.png.we Tom's Hardware is says "10nm" Alder Lake is being treated as a serious Intel technical advancement by Tom's analysis team. It can slightly out-game the current AMD units by using a selection of the fastest of all the Alder Lake big cores at once and overclocking the hell out of them until they overheat. It can "out compute" AMD by using a mix of Intel efficiency and performance cores and overclocking the hell out of them until they overheat. The Intel processor management software works, but needs some updating to fix a few minor bugs that showed up after launch. Repeat, the Intel processor management software is working fairly OK now and the nightly updates will fix the current minor bugs. Alder Lake is designed as non-homogeneous and the Intel Processor Monitoring software is key to using the complex non-homogeneous processor. As a core gets tired and the performance drops, it becomes a next level down core automatically. This software management means there are no substandard Alder Lake chipsets, as there is room at the bottom for the downgraded cores to be useful. As each processor cores get overly warm, they are deselected by the processor management software and the load is moved over to lesser performing cores. The huge chipset has enough backup cores that are needed to do this trick ongoing, permitting overheating to dissipate while maintaining overall performance. The only downside Tom's sees is Intel's very large power consumption and the BIG BIG HEAT LEVELS which inexorably leads to across the board BIOS throttling as the chipset cooling cannot keep up with the heat flux. Tom's recommends spending some extra Intel system money on a first rate liquid cooler system. Intel will be rated #1 on Tom's processor listing until the next AMD generation ships in a few weeks. This head to head bash up takes place at only 3 processors, with Intel lowering their price to 25% less than AMD while edging past AMD's performance, with AMD moving their selling price down to match the Intel price point. Lots of gamesmanship taking place here, with not a whole lot of across the board clear wins that will last more than a few weeks. Warehouse supply of these contested chips has instantly evaporated causing some pundits to accuse both of the competitors of doing a paper launch and a paper cost reduction with no inventory to back either up. My central point is both processor makers are ACTUALLY REACTING AND COMPETING AGAINST EACH OTHER and that is a good thing as they will cause each other to make real improvements at a greater speed ....... Intel's speed improvements and throughput improvements (at a stiff cost in power consumption and high processor heat, requiring a new motherboard loaded with state of the art very expensive DDR5 new memory and a needed water cooler) will remain, and Intel's package will be here all of next year to keep AMD honest. AMD offers the same level of performance using your existing memory and motherboard socket for just one more upgrade, then next fall AMD will go to a new socket and a new motherboard as they drop down to 5nm. Chiplet counts from both players are DOUBLING late next year, with better architectures and better throughput from each. Intel will triple their power draw from what they are selling right now and will require expensive liquid cooling on all desktop chipsets to keep heavy throttling at bay. Intel sees no issues with cooking things if the cooling system hits a hiccup, this has actually been the Intel way for a while now. So, if you don't have the recommended cooling solutions, it's on you, baby. Apple will double their processor counts also, while dropping down to 4nm and 3nm lithography levels. Rumored core counts of 32 and 40 cores are being floated by folks who claim to have seen them in TSMC factory packaged Apple chipsets. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/16/21 at 19:15:35 https://liliputing.com/2021/11/lilbits-windows-10-21h2-is-here-and-windows-11-learns-the-true-meaning-of-the-b-in-bsod.html Apple, MS, Intel and AMD are squaring up to take some bigger risks. The Blue Screen Of Death will reappear more and more on Wintel machines because MS is using all of installed PC's as their test group. MS knows there will be issues as what they are putting out "cannot be thoroughly tested" items against all PC constructions all at the same time. It will go out in a nightly update, fail, report back home to MS what went wrong and MS will trickle back out some fixes that will likely make some other issues happen. Can you say "cluster fook"? Just do a rewind of 3 years ago to see what it will be like to live through this again ...... Linux isn't happy because Intel and MS are pushing out some of this strange crap to fail in their Linux world, too. Mint is acting plumb defensively at the moment, holding back MS and Intel "advancements" until they become stable in the Windows 11 Linux Implementations. Linux Mint values stability, so it is staying back an entire release generation from all this whacked out stuff. Mint will implement the Win 11 improvements very selectively and very slowly after a separate run of Mint testing. Most of what Wintel is doing is trying to fix the current issues with Intel's latest processors and it involves 10-24% processor speed delays which frankly don't do much of anything for the older Linux crowd of hand-me-down machines and really aren't needed by them at all. It looks like all of Intel's small performance gains are now being absorbed by the slowdowns put in by MS as the required Intel security meditations. It is a whole lot of stuff to do just to prompt AMD to put in the next stair step to heaven. =================================================== Intel is moving off their dime finally. Intel has placed orders and taken shipment of some ASML 3-5nm Scanners. Intel is making waves and shooting for some significant processing advancements. Intel is back, and that is a good thing for the industry in general. Apple will soon be Intel's competitive focus because Intel has arguably equaled AMD at this particular point in time and both AMD and Intel will nudge ahead of the other in small baby steps going forward. But it is Apple that can break out into the commanding position with their 3nm 32 and 40 core ARM based chipsets currently in testing. Apple gets all of TSMC's full support (they pay handsomely for this service, BTW) so expect to see TSMC 2nm & 3nm trotted out in Apple products for the first year starting this upcoming year. AMD is making big moves in mainframe chipset construction right now and AMD is putting the Xylinx accelerator stuff in there as well. AMD is now considering Samsung 3nm as a possible interim step as TSMC has some serious 3nm yield issues that are being very troublesome at the moment. If AMD is going to compete in two years, they need a large volume of 3nm and 2nm chiplet wafers. ================================================== Confirmation comes on AMD moving several products over to Samsung's 3nm process as TSMC is not having "readily expandable success with 3nm" off their current lines. Plus, all existing 3nm capacity belongs to Apple for the next several years. Apple's needs come first, and Apple isn't getting all they want right now either. AMD going over to Samsung leaves Intel pretty much out in the cold for next year since AMD had priority over Intel going into this planning period. AMD is not giving up any TSMC allocations, they simply are tacking on Samsung as chiplet volume expansion as it is covered by their TSMC agreements as the documented "first alternate supplier". Samsung 3nm is a fairly firm and reliable source for a one-two year planning horizon. Count on TSMC to fix their order issues and to come back to AMD with a good proposal when they have the free capacity to build all the AMD chiplets that AMD needs. AMD is currently taking allocations for all the 5nm and 4nm AMD chiplets that TSMC can make, beyond the ever growing totals of what Apple needs with their new M1 Pro[ch8204] and [ch8204]M1 Max[ch8204] chipsets. It remains to be seen if AMD can mix and match chiplets from TSMC and Samsung on the same product socket board. If AMD can do this, then their current pain and suffering will diminish a lot accordingly. All manufacturers using chiplets rank them in 2-3 different performance groups, allowing for a core to degrade some and simply be utilized as the next step down (no scrap in this ranking system). |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/22/21 at 22:40:35 https://www.tomshardware.com/news/phytium-d2000-eight-core-for-desktops http://https://cdn.mos.cms.futurecdn.net/Je7vdDVQHCTwgP393dfdXm-970-80.png.we This new company mini PC has an ARMv8 processor and AMD Radeon RX 550 graphics People wonder why Intel keeps losing 1-2% market share every time anyone reports market share numbers while everyone else gains 5-8% during the same period. How is this possible? The answer is that massive world-wide compute industry expansion is going on in a steady fashion. The world market share overall pie is growing constantly with new users popping up all over China and India and Indonesia. AMD is growing a little better than the Industry Growth Average while new ARM and RISC-V processors are providing the bulk of the newly built processors making up the background growth numbers in a thin spread all over the place. Intel is shrinking steadily (relative to the market as a whole) as the brand new market players eat Intel's lunch for them. Look to see 8 cores at least double with the next downshift in lithography ........ Intel always only compares its own numbers to itself on purpose, so with a straight face Intel can saying they have market growth while Intel's significance and relevance continues to shrink along with Intel's real world market share numbers. It is a constant on-going lie that Intel has to provide to reassure the Intel stockholders ........ ARM v9 is coming next year, and with it come the first smells of a tipping point for x86 desktop dominance. Watch what Apple does with ARM processors to see the pathway of the future unfold before you. 32 and 40 ARM v9 and X1 cores are coming this spring from Apple ---- some significant processing power here that is overshadowing both Intel and AMD on the low end consumer side. Mediatek is doing something similar with some very strong ARM chipsets. AMD is providing their state of the art graphics to both Mediatek and to Samsung. Samsung is known to be doing runs of 3nm graphics chiplet wafers for AMD in exchange as nobody can find any loose TSMC wafer capacity for next year. =================================================== https://liliputing.com/2021/11/this-mini-pc-has-an-armv8-processor-and-amd-radeon-rx-550-graphics.html Wow, here is another one popping up ---- possibly the same main board in different cases. http://https://liliputing.com/wp-content/uploads/2021/11/d2000_03-768x378.jpg |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 11/27/21 at 06:28:29 Rumor Time These are still rumors until they actually start shipping. The ARM buyout by Nvidia is now very unlikely as 5 different regulating bodies have given a long list of stuff to Huang to answer and he is balking at providing any answers to any of them. So, he gets "no" votes by Britain, Brussels, China and the USA until such time he gives up satisfactory answers to each regulator body. Huang is now content to let the buyout clock run out and not spend any more time, effort and money on his aborted ARM Holdings takeover efforts. ARM is currently rolling out newly designed 5nm processors with even newer 4nm design specs being provided to the world as tryout parts for the next 4nm generation and the following 3nm generations, with Mediatek acting as the prime "first implementer" of the newly implemented ARM 4nm standard designs right now. ARM/Mediatek 4nm is now looking at a new doubling on the core counts with a fresh 30% performance upper along with a memory capacity doubling bumper due to use of the new TSMC "vertical stacking" and dense packaging techniques. This ships starting first quarter next year. Ditto again for a 3nm speed bump in 2022 ...... For example, 8" Fire Tablets might pick up 30% faster processing with DOUBLE the systems memory on all the base units within the next 2 years. Amazon contracts with Mediatek far in advance and Amazon only upgrades once every 2 years. This upcoming Amazon Fire processor bump (or mebbe the next one) will possibly memory and speed lap my old desktop PCs ...... in a silly cheap tablet no less. ::) Graphics processing from ARM is getting better too, but those who want really good graphics (Samsung for example) are doing deals with AMD to get AMD Radeon ray tracing graphics for their devices. =================================================== Their toes are being nibbled at by odd looking Chinese ducks ....... For $17, RISC-V is selling an equivalent to the original Raspberry Pi device. But who would want a 4 years out of date copy of the original Pi? |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/03/21 at 01:57:31 Now for some rumor becoming cold facts. https://liliputing.com/2021/12/sifive-performance-p650-risc-v-processor-revealed-as-a-challenger-to-arm-cortex-a77.html This says that RISC-V is now only one-two generations back compared to ARM. https://www.ftc.gov/news-events/press-releases/2021/12/ftc-sues-block-40-billion-semiconductor-chip-merger This says that the Federal Trade Commission has sued NVIDIA to STOP all their log jamming bullshite, stating that NVIDIA will NEVER be approved to purchase ARM. NVIDIA will have to pay their cancellation fees to ARM and IMMEDIATELY STOP stalling the development of all of ARM'S new processors for all of ARM's other customers. Any NVIDIA tech that got crammed into ARM's designs unwillingly by NVIDIA during this grey period will get a perpetual license from NVIDIA and NVIDIA will pay ARM full license fees for any tech items NVIDIA lifted from ARM. ARM and the phone industry in general were damaged by NVIDIA, and the FTC will try to see that NVIDIA picks up the tab for that damage in as much as possible. What the FTC can't fix is all the yardage that RISC-V made up while ARM was all log jammed up by NVIDIA. Many companies moved away from ARM in reaction to Jensen "Benny" Huang and instead have firmly committed to RISC-V and that RISC-V commitment remains and all the new RISC-V processors that are under design by multiple companies will be completed and shipped next year. ARM Holdings may never really recover from this convoluted mess in the current generations of processors ----- look to see ARM expedite some new generations of processor designs to pick their market share back up again. NVIDIA responds saying there is no basis for the FTC suit --- take us to court if you want to. NVIDIA specializes in legal quagmires, so FTC would be foolish to do so. Some would say that blowing off the Federal Trade Commission is a very poor idea, as payback is a right biatch when it come for you. Next, new regulations can become pointed weapons if you piss the government agencies off badly enough. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/12/21 at 23:28:01 Intel vs AMD AMD is year on year moving on down the lithography chain at TSMC. AMD is hitting each announced step in their roadmap like clockwork. Intel is also hitting their roadmap steps, but is having a far bumpier ride in doing so. Intel is still sucking down twice as much power as AMD and is still requiring a brand new motherboard per each step of the way. AMD is becoming preferred in the data center market as AMD's progress steps work across the board making large improvements in data center that Intel cannot match. Intel now has their Intel 10nm working in volume production (they call it Intel 7, but it is really the old Intel 10nm super fin). Next year everybody goes down a lithography step. Intel still tries to claim parity with AMD using some relatively extreme power sucking tricks, but hey, the tricks are working and Intel is moving on up in the processor comparison charts at Tom's Hardware. As of Alder Lake, "Intel is back" as a competitor in the desktop and laptop space. However, AMD has improved its supply of chiplet wafers at TSMC by roughly 20% and is taking more market share from Intel accordingly in all nodes that are not Intel Alder Lake based (2.4% against the total x86 market just this last month). AMD is now using 3nm, 4nm, 5nm and 7nm TSMC wafer allocations to make this market share progress. Apple and Intel are beginning to squeeze in on the AMD's chiplet allocations by their own increased demands, so any new allocations will likely go to whomever is willing to pay more for them. Intel has hatched up an impromptu plan to buy up all of AMD's wafer allocations right out from under AMD and to put AMD out of business. AMD has countered this trick by paying for their allocations years in advance by putting in fully pre-paid production orders. TSMC is beginning to insist on everyone doing these firmly scheduled and fully pre-paid allocation production slots from everybody. TSMC likes the AMD plan for production scheduling, so everyone will do that starting now. If you don't like it, Mediatek or AMD will buy up your now unscheduled production slots and that is the harsh new reality in today's fabless chip production. We will see if the new USA plants being built in Arizona by TSMC and Samsung can alleviate the "production wafer war" which is currently building up as we speak. =================================================== I have said several times that TSMC is treating Intel like a competitor fab not as a customer. TSMC verifies this "competitor fab" status by telling Intel's CEO on his last trip just this past week that Intel has to pay up front for 100% of the materials and all run charges up front whenever Intel schedules a run, not when Intel takes delivery ---- and that all scrap produced on an Intel design run belongs to Intel. Take it or leave it, these are the terms. Intel is not a happy camper right now ........ Intel had often messed up initial runs of new designs as they kept designing things for their own Intel processes instead of designing things from the get-go according to the TSMC ASML standards. Obviously, TSMC does not really want to make chips for Intel having seen first hand how Intel repeatedly screws up and Intel then stiffs their suppliers and their business partners over the years. TSMC is simply acting to protect themselves by having these Intel restrictions put in place up front. AMD will buy up any slack that Intel puts into the supply chain, followed by Apple, Mediatek and several others. Unused production allocations will get resold promptly and Intel goes to the back of the line to schedule some new TSMC runs again. TSMC lists Intel all the way down their list of preferred customers, all the way down at position #11 on their list of preferred customers. TSMC has only committed (by getting paid their full price up front) to building some selected Intel graphics processors for Intel at this point in time. These graphics processors were the ones developed by TSMC on TSMC's production processes, so they should work without causing big scrap charges to anybody's bottom line. This may change, of course, as Intel starts buying and using only industry standard ASML process lines as their bread and butter production equipment. =================================================== AND YES, THE INTEL ALDER LAKE FAMILY REALLY IS KICKING AMD'S BUTT FOR BOTH PERFORMANCE AND PRICE AT THE MOMENT ...... And yes, this is intentional and active price fixing on Intel's part as these new chipsets are larger and more expensive to make but are being sold for less. http://https://cdn.mos.cms.futurecdn.net/GmAA54PHYNWbbbafJgKgHf-970-80.png.we |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/22/21 at 03:37:10 https://liliputing.com/2021/12/lilbits-ces-2022-will-likely-be-subdued-as-companies-pull-out-over-covid-19-omicron-variant-concerns.html The new strain of Omicron Covid has likely killed off a physical CES show again for this year as major presenters are now dropping the show right and left again to protect their key people's health. They still may have the show for the now fashionable liberal anti-vax crowd to show up and get all infected with Omicron Covid variants, =================================================== https://www.tomshardware.com/news/amd-leads-growth-in-ic-industry-2021 AMD Leads Growth in IC Industry for 2021, Intel Shrinks Alone The whole year of business roll ups are beginning to come in, and Intel's last minute blip or single well timed (and well price supported) ray of Alder Lake performance sunshine does not make up for a generally gloomy total 2021 year picture for Intel. According to IC Insights, the top 17 semiconductor companies are bound for a combined $460 billion figure in Integrated Circuits (IC) sales for 2021. All 17 companies have now surpassed the $10 billion mark, but like in most markets, the top companies typically hold the vast majority of sales. Intel, Samsung and TSMC are projected to account for around $210 billion of the industry's total sales revenue, and Intel seems to have no chance but to hand the semiconductor crown to Samsung. In fact, Intel is the only player in the top 17 not posting a positive gain for 2021, instead posting a 1% contraction in sales. This is -11% effective to the industry average growth of +10%. That translates to about $800 million less money in the bank. The company has been facing supply issues for longer than other manufacturers, mostly related to the various delays with its 10nm and 7nm products. The company even started porting some of its semiconductor designs (e.g. Comet Lake) to less technologically-advanced production nodes. The fact that the company took so long to ramp up its 10nm SuperFin technology also meant that a highly competent AMD could wreak havoc with Intel's portfolio through solid execution — and more than a little help from TSMC's expertise. AMD itself, which has been executing on its roadmaps with a cadence that once would have made the old Andy Grove at Intel proud, AMD is expected to post an incredible 65% revenue growth compared to its 2020 results, going from $9.7 billion in 2020 to $16 billion, and breaking the $10 billion milestone. NXP and Analog Devices will also seemingly manage to hit that mark. MediaTek comes in behind AMD with its also tremendous 60% growth, buoyed by high-performance Arm designs and by the company's 5G portfolio. So, 10nm super fin (alias the BS renamed "Intel 7") simply came too late and too little to help Intel keep its overall ranking in the industry. This also brings up the lack of current general lithography supply issues within Intel again as there are not enough 10nm super fin lithography lines to run all the chips needed to move Intel forward in any sort of sufficiently fast fashion. A Deep Breath and a pause for reflection. This next image is one day old. This is a tired somewhat ragged looking Pat Gelsinger proudly pointing at a 3 layer thick combined 10nm & 14nm super fin processor sample, one built using Alder Lake technology that is sucking in all sorts of power to just barely beat AMD's current products that run twice as economically on much less power consumption. This is a picture of "Intel is now trying to spend AMD into the ground" in other words. This combined lithography, 3 stacked layer monster is PHYSICALLY HUGE (over twice as big as AMD's M4 socket product and it is pushing old 10nm and 14nm technology strongly into places it has never been taken before. BUT it is working at a very uneconomical power level ....... and Intel can apparently afford to build a fair quantity of them at a partial loss right now. It also requires a new motherboard with an equally gigantic socket and a equally capable supersized cooler to cool the beast down, as its net performance advantage is slim when cold and that advantage can quickly disappear due to the processors overheating. Alder Lake is still winning on paper against AMD at the moment, but it is taking up a LOT of Intel price support $$$ to do so. http://https://static.techspot.com/images2/news/bigimage/2021/03/2021-03-23-image-22-j_1100.we Pat is looking and acting sort of desperate somewhat right now because if he can't get his hands on a chunk of the $80,000,000,000 in Biden Build Back Better Bucks real soon his Intel house of price supported cards is going to start to fall down around his ears again ........... :P If he does get his hands on the Biden money he can limp along pretty good until he needs his next hand out, as he has spent his own money like a true liberal democrat (i.e. spent CAPITAL EXPANSION money he does not really have) for most of this past year. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/22/21 at 13:55:33 Good news ---- Intel is finally taking delivery and installing "two" state of the art ASML scanners, but they are for year before last's 5nm lithography level though. And not enough of them are being installed to support all of what Intel needs to be building right now at 5nm. Bad news ---- Build Back Better has just died in Congress and Intel won't ever get their Build Back Better Bucks (sorry). Meanwhile, AMD has been rolling down to 6nm headed firmly towards 5nm and 4nm and soon will start up a heavy 3nm production flow on their video card system chips. I find it odd that 3nm is good for video card processors, but is not good for CPUs (unless you are Apple, that is) ::) =================================================== https://bits-chips.nl/artikel/asmls-backlog-doubled-in-q2/ This is an important read for you ---- ASML was sitting at over 3+ years of active delay on their total waiting list ........ but now ASML is now being told by their own Dutch government to take drastic action to double up the production rate & numbers of their most modern scanners, and to do it ASAP. This is their own government telling them to do this on a national priority basis as all of the NATO Alliance depends on processors from state of the art ASML production machines for their technical and military survival. Funding for this extraordinary floor expansion at ASML is being provided by the US and others, these are the same others that see China flying flights of attack bombers over the Taiwan TSMC plants as a direct threat being made to their own country both economically and militarily. Needless to say, these are going to be brand new modern EUV ASML scanners that will not be installed in Taiwan which is geographically totally and completely subject to the whims of the Chinese military machine ....... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/26/21 at 05:26:46 AMD and Intel are releasing new "rumors" about their next generations. AMD talks a bit about some future thing that has a casual or real basis, and Intel will immediately pop off some "far future" plan that 1) is shortly overwritten by their later waves of released BS information 2) is never intended to be taken seriously as Intel does not know what they really plan to do that far out. AMD is now talking about solid plans for 30 times more processing speed using Xylinx tech combined with AMD 64 bit x86 tech. This presumes TSMC continuing on down the lithography chain, which is at 7nm drifting down to 5nm. These are well documented "mainframe" style plans, plans that you must do after announcing them. Last time AMD talked about this sort of stuff, it was for a 20x improvement level which AMD then hit and exceeded on their claims as TSMC 7nm matured. AMD designs all their stuff for MAINFRAME processors first and foremost, then adapts the more capable chiplets for their consumer & gaming machines. Mainframe users want lower power consumption (saves money on utilities and on data farm cooling) and ever increasing processing speed. These mainframe priorities then get reflected through the rest of the AMD product families. AMD chiplets are already twice as energy efficient as Intel tiles, with both players offering about the same current processor speeds (Intel just uses far more power and cooling $$$) to get there. AMD offers far cheaper motherboards with standard A-4 sockets that continue to be able to be upgraded with just installing a new processor and upgrading the board BIOS. AMD will upgrade to an A-5 socket soon, which mean buying a new motherboard will then last for approximately 10 years before getting replaced with an A-6 socket motherboard. Intel intentionally churns their entire world of stuff every year lately---- this means lots of new stuff is force sold to Intel fanboys whether needed by them or not. Intel motherboards are now costing twice as much as AMD motherboards and that trend is getting more and more noticeable as YOU CAN"T JUST GO OUT AND BUY THE LATEST STATE OF THE BRAG INTEL MOTHERBOARDS RIGHT NOW (various chip shortages, you know). ================================================== MindShare, Amazon and other major processor sales centers that supply users that build their own systems continue to see AMD latest and greatest selling through in far larger numbers compared to Intel. Intel is currently being hampered by lack of new motherboards to hold the new huge Alder Lake chipsets and MS itself lacks the Windows OS optimizations to properly support the new Alderlake processors. Attitude in the Computer press seems to be that "in 4 weeks AMD comes out with their next generation of stuff, talk to us after that stuff gets rolled out and we will see the 2022 winners." I think Toms Hardware has called it, in 4 weeks the stuff will be all neck and neck again. Mainframe and workstation boys will prefer AMD as it is more stable, sockets last longer, it requires half the power and half the cooling system that Alder Lake would require and systems memory required is cheaper and easier to buy. People buying pre-made boxes will only see the same old Intel stuff sitting out on the shelves until all existing stocks are all bought up ....... (Intel still has that strong of a legal contractual lock on the box builders). |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/29/21 at 06:02:16 https://wccftech.com/intel-core-i5-12400-beats-the-core-i9-11900k-rocket-lake-flagship-in-leaked-gaming-benchmarks-at-less-than-half-the-price-while-consuming-less-power/ Intel Core i5-12400 $180 US 6 Core 4.4 GHz CPU Beats The $500+ Core i9-11900K 5.3 GHz CPU In Gaming Benchmarks The Intel Core i5-12400 will feature a 6 core and 12 thread design and will comprise only of Golden Cove (P-Cores). The Intel Core i5-12400 Alder Lake CPU aims to be one of the best gaming chips, not only within the 12th Gen family but in general. We have already seen several benchmarks of the chip where it runs on par with the Ryzen 5 5600X and even beats it in several cases while consuming lower power and running cooler. I used to say that Intel cheated on benchmarks, then I began saying that modern benchmarks needed to be re-written so that Intel couldn't spoof them quite so easily. Now we have a $180 Intel Alder Lake i5 chip with some interesting claims of equaling and beating the Ryzen 5 5600x and doing this with all of its Alder Lake efficiency cores lasered off with this brutal gelding being done simply to cut down on the huge amount of power this HUGE Intel circumcised monstrosity still wants to consume. Now here is the real rub --- the $180 Intel circumcised Alder Lake struggles to beat a Ryzen 5 5600x but the wild arsed results Intel is claiming for this $180 i5-12400 chipset exceeds the gaming benchmarks of their own $500+ Intel Core i9-11900K 5.3 GHz CPU....... wow :o ::) :o Yet Intel also still claims the i9-11900K at 5.3 ghz beats all comers by a wide margin ....... (very strong smell of baitfish baking in sunshine on the aluminum boat seat is going on here). Intel marketing is simply beyond my ability to follow ---- and most PC gamer fans seem to think likewise because they are NOT buying very much Intel Alder Lake at the moment. The customer base seems to be waiting for AMD to come "one up" Alder Lake with something that is totally real that can be both independently tested and readily understood. ;D And yeah, you know it is coming, soon real soon ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/29/21 at 21:32:23 China is stacking old style primitive large lithography stuff up on three layers deep stacked daughter boards (something they learned from watching Intel, I think). All of the limited support documentation is in Chinese, and it is FAR FAR less than a current Raspberry Pi at this point in time, but it came on quickly, very quickly ...... http://https://liliputing.com/wp-content/uploads/2021/12/lichee-rv-dock_01-651x500.png http://https://liliputing.com/wp-content/uploads/2021/12/lichee-rv-dock_03-780x481.jpg The Sipeed Lichee RV is an inexpensive computer-on-a-module featuring a single core 64 bit Allwinner D1 RISC-V processor, 512MB of RAM, and a microSD card reader for storage. First announced in November, it’s a module that you could theoretically use as a standalone computer thanks to an onboard USB-OTG port, it’s missing a few features like a display output. But now Sipeed is selling a tiny, affordable dock that gives the little computer HDMI and USB Type-A ports, among other things. The Sipeed Lichee RV module sells for $17 and you can pick up a dock for $5 (or $8 if you want a model with built-in WiFi and Bluetooth support). You can also save a buck by purchasing a $21 bundle that includes a LicheeRV + dock. ;D Wow, a clone of the original single core Raspberry Pi power level with modern I/O and M-2 storage coming from China for $25 complete with finished I/O, substantial storage and Ubuntu support. The Chinese do not lack the get-to-it-tive-ness needed to overcome obstacles and to compete. Makes you wonder what Allwinner will be shipping internal to China later this year, and how many short years it will take for the Pi Brits to feel threatened by this very rapid RISC-V development and growth enough to actually go significantly upgrade their Brit stuff again ........ http://images/prettyPhoto/light_rounded/btnNext.png |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 12/31/21 at 06:34:37 CES AMD "pre-announcement" RUMORS 5.0+ ghz processor speeds are achieved off the production stock cooler set ups. Power consumption still be less than half that of Intel. Throughput will be neck and neck with Intel again. Memory for this generation is far cheaper than what Intel requires. Processor and socket are still less than half the VERY LARGE SIZE and huge power draw of Intel's Alder Lake processors. =================================================== Reality check on them rumors listed at the top of this post ......... Yep, 5+ GHZ achieved, got double the memory on top of the existing layers (3 layer deep) Package is same size area-wise but has slight difference in pin out configuration. Looks like this ........ can you see them three thin 6m full sized layers stacked neatly on top of each other? http://https://cdn.videocardz.com/1/2022/01/Rembrandt-APU-768x591.jpg Supposedly this 6nm triple stack's performance is neck and neck or slightly better than what Intel Alder Lake can do (with Alder Lake burning over 2x the energy use from your device and demanding 2-3 times the cooling energy consumption --- but hey, you will need to wait a week for the real world testing results from Toms and ArsTechnica and the others as AMD does not send out a press package complete with all the tests pre-done by some Intel paid specialty house off of some special binned chipsets conveniently run on a freon cooler rig. AMD simply says "go buy an upcoming Ryzen 6000 APU series chipset at retail inside a real competitor product and then test it and tell us what you think about it FOR REAL. :P REMEMBER, Intel's latest and greatest 10nm and 14nm 3 layer stack up still looks like this. Why is Pat Gelsinger so proud of that GIGANTIC HUGE POWER-SUCKING BLOATED PIG OF A PRICE-SUPPORTED INTEL OVERBUILT ABORTION I certainly do not know. http://https://static.techspot.com/images2/news/bigimage/2021/03/2021-03-23-image-22-j_1100.we Alder Lake is not really selling through very well and now you know why. Intel has just announced new plans for TSMC to build entire production generations of their chipsets for them and now you know why about that too. Would you buy a fab from this guy using this level of technology ????? ================================================== I watched the CES keynotes from AMD and Intel ----- I came away amazed that AMD is bluntly saying the 6000 APUs just out from AMD will outperform dedicated NVIDIA laptop graphics cards plus the best current Intel processors that are shipping out in this quarter. 30+% more powerful and 30+% more efficient --- this would be hard to beat by Intel or by ARM. Frankly, the news from AMD is so good I have to say "Let's wait to see what ArsTechnica has to say about the first finished units before getting all excited". http://https://liliputing.com/wp-content/uploads/2022/01/zen-4_03-572x500.jpg BUT THERE IS MORE ..... Ryzen 7000 desktop chipsets will offer similar or slightly better uptick results yet again in the second half of this year. To try to really compete against Ryzen 7000 on the desktop, Intel will HAVE to have TSMC up and running on large production lots of newly designed Intel chipsets, or else have dedicated their very few newest Intel Dutch 5nm ASML scanners 100% to run these brand new designs and have them all working correctly right off the bat. Some really big risk levels in this for Intel, in my humble opinion ....... =================================================== I just re-watched the Intel CES presentation again ----- and I was not deluded or mis-remembering. Intel said NOTHING and used a whole lot of vague generalities to get there. AMD had 6nm hitting now and 5nm starting in about 5 months, in both cases quoting throughput improvement specs and concrete expectations along with current draw expectations. Aside from some vague generalities, Intel has nothing concrete to communicate at CES. I question if Intel has even got a firm plan on what they are going to be building and where they are going to be building it six months from now, much less any form of sampled processors. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/04/22 at 13:31:49 Intel had been pushing loud claims of 25+% improvements in their laptop performance suddenly jumps their claims to 40+% when faced with 30+% improvements in Ryzen 6000. This is the same war of lies that Intel wages all the time. Simply run your tests hotter and you too can beat AMD's best. Trust your specialty house and their Freon cooler rig to keep it all together. For how long does not play in any Intel testing as they never wait around for the inevitable BIOS throttling to take place. Just quote all performance numbers off the big cores and take the energy usage off only the little cores ........ ::) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/04/22 at 23:03:37 Claim and counter claim have run their course at CES yet again ...... Intel is still in the game and is now desperately scrambling to get TSMC to run them a low lithography family of chipsets so that Intel can survive for another year. AMD's roadmap is simply implacable, and AMD has backup production scheduled at Samsung if TSMC cannot make all of what AMD needs. Intel is competing now on all fronts just as hard as they can. But Intel is behind on lithography and Intel's old partners like MS are suddenly showing up over in AMD's corner ....... Intel cannot simply run hotter and hotter and use this to claim to be faster any longer. A day of reckoning has come, motherboard charring is taking place under Intel's massive chipsets. Cooler tech to support what Intel needs for cooling is not there, not in anything you can take home with you anyway. But credit to Pat's boys is due too, they still keep on trying just as hard as they can no matter how off base they may be. Lies during CES are not new from Intel, but this complete lack of any new anything "factual" is somewhat clear to those who choose to look at it. Intel is beyond stressed, it is completely punch drunk. Hard Fact: Intel is shrinking in an expanding marketplace. TSMC and Samsung are both larger and more important than Intel and none of Pat Gelsinger's "envisioning" is going to change that. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/04/22 at 23:47:47 https://www.youtube.com/watch?v=svRDkmmjvuM "Stay tuned, more Intel excitement is ahead." Really? I bet this latest BS wave gets sorted by the reputable computer press within 2-3 days from now ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/06/22 at 10:01:53 https://www.theverge.com/2022/1/6/22870060/intel-announces-5-5ghz-capable-12th-gen-cpu After teasing the news earlier this week, Intel has officially announced a new 12th-gen desktop CPU that’s capable of boosting to 5.5GHz on a single core out of the box. The news came during Intel’s CES 2022 presentation, when it said the KS-series desktop processor would be shipping later this quarter, but there was no mention of pricing. The full name of the chip is the Core i9-12900KS, and Tom’s Hardware notes the “S’’ suffix suggests it’ll be a limited edition. As well as hitting a maximum of 5.5GHz on a single core (a 300MHz improvement over what we saw with the i9-12900K), an onstage demonstration of the new CPU powering Hitman 3 showed the processor running at a sustained 5.2GHz across all its performance cores. No details were given on what sort of power draw or cooling solution was needed to achieve these results. The processor was announced onstage during Intel’s CES presentation. Unfortunately, it doesn’t sound like you’ll be able to buy Intel’s 5.5GHz CPU for your own rig anytime soon. Intel executive vice president Gregory Bryant noted only that the CPU would be shipping to “OEM customers” during the presentation suggesting that, in the immediate future, you’ll have to buy a prebuilt PC to get your hands on one of the chips. OK, most Alder Lake chips have a single good core that can hit 5.2 GHz for real, just like Alder Lake normally can do. One out of 100 processors has a single core that can go up to 5.5 GHz on these specially sorted and binned processors, but only for a short duration (only seconds in duration), but only when using an Intel "recommended cooler system" is used and the processor is stone cold on a chilled cold start up and only when using a special Intel recommended power supply (as a very large wattage of power is required to go that fast). These specially binned Intel processors supported with special cooling requirements and over-sized power supplies will only be provided to certain favored Intel OEMs who will post their testing results off of them so Intel can make more "verified" exaggerated claims of their Alder Lake prowess. ;) Standard Intel smoke and mirrors is going on to a degree, in other words ....... =================================================== Linus Tech Tips has some words to share about that Intel "recommended cooler". https://www.youtube.com/watch?v=UPdpS_6ln-o ........ and the real progress AMD has made just these last few weeks. https://www.youtube.com/watch?v=kGHbrUOoKjI There will likely only be one midrange 6nm AMD chipset supposedly, as AMD 5nm will roll up to their next step in the road map very very shortly, especially shortly if Intel continues to keep the big big pressure up with Alder Lake processors they way they have been doing. AMD has bigger guns already sitting in the armory and a whole generational sized lithography progression they can quickly uncork to compete more effectively with Intel. Intel, on the other hand has two 5nm ASML lines they just took shipment on ....... ::) On each side of this competition the players can't do everything across the board all at the same time, so look to see each one do what counts the most for each one of them. ;D |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/07/22 at 04:37:18 https://www.youtube.com/watch?v=-EogHCFd7w0 Linus Tech Tips says Win 11's big and little core scheduler and Intel big little processors constitute a compelling win right now and AMD needs to bring forward their next competitive (big little) processor generations immediately. AMD can no longer intentionally stretch out its progress pipeline because Intel "requires no additional response from us" right now. Intel requires a response alright, RIGHT NOW !!! AMD certainly can't complacently follow its 2022 roadmap as is because the next two stair steps in that roadmap are already partially overcome by events. =================================================== There is one other cause for "sticking to the schedule" delay, however ------ you must pre-schedule your TSMC production slots and pre-.pay for your production materials costs when you do the scheduling. Sunk costs constitute a pretty good reason to follow the plan, but to also adjust that plan as much as can be done at this stage of things. AMD has good relations with TSMC, so they may have some wiggle room here --- especially if TSMC is taking some image hits from Intel's unexpectedly good performance on this topic. Once again, look to see TSMC treat Intel more like a competitor Fab trying to offload a capacity issue rather than a simple pay for a production lot customer. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/10/22 at 17:32:52 https://liliputing.com/2022/01/pbp-vs-tdp-intel-changes-power-consumption-starting-with-12th-gen-chips.html PBP vs TDP: Intel changes power consumption definitions starting with 12th-gen chips by BRAD LINDER 01/10/2022 PBP vs TDP: Intel changes power consumption starting with 12th-gen chips. For years Intel has used TDP or “Thermal Design Power” to give a rough approximation of how much power a processor uses. While TDP actually measures how much heat is generated as a way of letting PC makers know what level of cooling is needed, it also gives users a hint at a chip’s power consumption… but over the years that hint has gotten more vague since chips can greatly exceed their TDP thanks to Turbo Boost and similar technologies. So starting with its 12th-gen “Alder Lake” processors, Intel is no longer providing TDP figures. Instead the company has replaced that figure with a new “Processor Base Power” or PBP number. What’s the difference? Honestly, as far as I can tell, nothing that makes sense. But the new name gives us a better idea of what the number is actually trying to measure: the average amount of heat that needs to be dissipated by a processor when running at its base frequency. More importantly though, Intel is also now providing a few additional numbers, including the “Maximum Turbo Power.” This should do a better job of letting you know not only how much power your computer is using when running at its base frequency, but how much it will use when running at higher frequencies using Turbo Boost technology. Processor Base Power (PBP): This is basically the new name for TDP, and measures the average level of heat generated during high-complexity workloads while a chip is running at its base frequency. Maximum Turbo Power (MTP): This figure represents the amount of heat generated at maximum Turbo Boost frequencies over “sustained” periods of longer than one second. Formerly known as cTDP Up, this number will be higher (often substantially higher) than the PBP. Intel debuted the new PBP and MTP nomenclature when the first Alder Lake chips for desktops launched late last year, but now that the company has also launched 12th-gen mobile chips, it’s particularly helpful to have the new Max Turbo Power figure because it will help figure out whether a laptop would benefit from active freon cooling or if it can run passively and/or give you a better handle on the chip’s expected impact on battery life when running resource-intensive tasks. Most Intel Alder Lake mobile product descriptions also include a “Minimum Assured Power” figure, which Intel describes vaguely as an updated name for what used to be called cTDP Down, or the amount of heat generated by a chip that’s been “dynamically adjusted to a desired system performance and power envelope.” In other words PC makers will have the option of underclocking/undervolting chips for some systems such as thinner, lighter, and potentially fanless computers and this figure gives an idea of the power consumption/thermal performance of chips running in this mode. There’s also now a “Maximum Assured Power” figure which suggests Intel will allow PC makers to overclock chips as well, but that figure doesn’t seem to be widely reported in the Intel Ark chip database. Intel throws out a set of new "non-definitions" of Intel input power draw numbers that totally hides the fact that Intel can now suck energy and throw out HUGE amounts of processor heat like nothing that was ever seen before the recent Alder Lake processors. Intel stock basic power draw is easily twice what AMD requires to do basically the same job. But that is just Intel stock basic power, Intel also has several turbo modes and even some much higher MAX TURBO MODES that can go double base power again and can to up to 4x higher than the Intel stock basic power draw in MAX TURBO MODE as the complex MS system balancing software sees any chance to use these any of these 4 various Intel hyperclocking schemes. THE TOTAL EFFECTIVE COOLING SYSTEM CARRYING LOAD in your hardware unit become the real output limiter for an Intel Alder Lake Processor. There is no such thing as cheating any longer, Intel and MS will recognize any chance to go faster just as long as they can dump the heat produced to do so. =================================================== Follow up on 1/25/22 The first laptops with Alder Lake are shipping this week. Performance and power draw numbers are being reported as: Processor Base Power and Maximum Turbo Power to indicate that these are chips that will use between 45 and 115 watts. For the most part you’re only going to see power consumption spikes up to 115 watts for very brief periods, but under load you can expect these chips to consume as much as 85 watts for an extended period of time. And that’s just the processor. |
|
Title: Re: AMD & others --- post AMD dominance in 2020 Post by verslagen1 on 01/10/22 at 20:32:28 Hey OF... there's your room heater right there. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/11/22 at 03:56:28 ::) Let's see ...... purchase price over $1600 and then you pay electrical utility rates for your electrical resistance heat, plus you pay for an internet service, plus pay a Microsoft tax every other year ....... no, no thank you |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/18/22 at 05:55:46 https://www.pcgamer.com/intel-ceo-says-amd-is-in-the-rearview-mirror-and-never-again-will-they-be-in-the-windshield/ Intel CEO says AMD is in the rearview mirror and 'never again will they be in the windshield' Intel's CEO Pat Geslinger gives encouraging Christmas greetings to his Intel troops who are still suffering from being laid off and degraded with no raises for several years by his predecessors. http://https://cdn.mos.cms.futurecdn.net/7SVVnapW2EAMhEbd5XwpJ9-970-80.jpg.we In it, Pat gives out his gut feelings on the next few years that supposedly will contain 4 waves of Intel manufacturing upgrades (some rented from TSMC, some purchased outright from ASML). Pat simply says that this is the low inflection point, the very lowest point in Intel's decline and from now on things should get better and better. If Pat Geslinger's house of cards holds up, he is likely correct in this estimation. Things are still somewhat shaky, especially with Joe Biden being too weak to move forward with his Build Back Better Plan but the deals and government bail outs still seem to be likely to happen eventually so I guess Pat is telling his troops the truth as he currently sees it. Intel's Alder Lake gives Pat his first technical leadership position for at least part of 2022, but AMD will take that leadership back when they roll down to Ryzen 7000 later this year. AMD requires no bail out bucks, which means that they will forge forward no matter what Biden can promise to give to Intel. Intel is currently running on brown vapor fumes and using some really extreme hyperclocking to brand new VERY extreme levels right now. Even so, Intel Alder Lake is actually up on top in the processor rankings right now ....... and if Geslinger gets his stream of Biden bucks he will likely trade places back and forth with AMD several times over the next few years. Historically speaking, an ascendant Intel always keeps AMD around just to keep monopoly claims from bothering Intel too much ....... as Intel freewheels around acting just like a monopoly. ================================================== An Eventful Week for Intel and others ...... Intel declares they are back on top again. Intel declares another 20 billion dollar new dual plant site in Ohio. Joe Biden is quick to claim credit for this imaginary progress for his administration (these are the same people who cannot pass the funding for what Intel is hanging its neck out so badly to do). NVIDIA declares they are a bigger American chipset supplier than Intel actually is and announces their plans to move CPU compute power over inside their big big graphics card sets. Yes, NVIDIA already has the tech to do this tomorrow AM if they choose to release the drivers to do so. NVIDIA already does this in mainframe world so the tech is not new to anybody. See AMD move to do likewise if NVIDIA does indeed move to do this. FTC follows through with their lawsuit to stop NVIDIA from taking over ARM Holdings (located in British UK). This puts the UK, America and Belgium against the takeover with China saying they will not give approval at this time. China is now doing the "will not give approval at this time" thing to the AMD and Xilinx merger now. Being strictly an inside the USA affair, technically it is none of China's business, but Joe Biden will apparently continue to allow China to be the sole (and only) obstructionist in this AMD / Xilinx case. Why ??? ...... because Joe Biden and his current administration is that weak and stupid. AMD quietly rolls forward with existing plans (and quietly puts out some increased requirements for some much better coolers for their biggest chipsets). AMD has TSMC both build and package all their new chipsets in their most modern plants in Taiwan, so do not be surprised when the new finished processors and cooling systems suddenly pop up in the world wide marketplace, quietly and without a lot of fuss. These new processors will be built with an improved cooling system in mind, as the ability to dump the heat made by the chipset actually makes up the "upper limits" that you can claim for that processor. Both AMD and Intel are top end limited by their cooling systems, so look to see a cooling system war quickly develop as both sides see the importance of good cooling. http://https://cdn.mos.cms.futurecdn.net/3QkWCdgG8dYUYnZhGDsNd4-970-80.png.we ==================================================== https://www.tomshardware.com/news/intel-core-i9-12900k-vs-ryzen-9-5900x-5950x Tom's ranks Intel as a "4 out of 5 categories" winner in this latest showdown comparison. AMD needs to bring forward their most modern biggest 5nm "big little" gun to deliver a big win across the board. Being unable to do so by the end of this year means Intel simply rules uncontested using their freakishly high current draws and all the various Intel hyperclocking tricks. The Intel hyper-clocking tricks will soon become the industry standard and AMD will be sidelined again as a stick in the mud "also ran". |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/22/22 at 20:01:11 MM, pay attention Potential daughter pc ....... https://liliputing.com/2022/01/morefine-m6-is-a-phone-sized-mini-pc-with-celeron-n5105-crowdfunding.html This is a little computer that runs Windows for only $256. Power comes from a 2 amp phone charger and it can support up to two 4K/60Hz displays when using the HDMI and USB-C port at the same time, thanks to USB DisplayPort Alt Mode functionality. The Morefine M6 is a computer with an Intel Celeron N5105 quad-core processor, up to 16GB of RAM and support for up to two SSDs, and support for Windows 11, Ubuntu, or other Linux-based operating systems. It’s also pocket-sized, measuring 155 x 80 x 19mm (6.1[ch8243] x 3.1[ch8243] x 0.75[ch8243]) and weighing 195 grams (6.9 ounces), making this mini PC about the size of a modern smartphone (or more like two stacked atop one another since it’s a bit on the thick side). First announced in October, the Morefine M6 is now available for pre-order for $256 and up through an Indiegogo crowdfunding campaign. http://https://liliputing.com/wp-content/uploads/2022/01/m6_08-780x474.jpg That represents a 30% discount off the estimated retail price, which means you’ll probably end up spending closer to $355 if you want to pick one up after the crowdfunding campaign ends. If everything goes according to plan, the Morefine M6 should begin shipping in April, but it’s worth keeping in mind that things don’t always go according to plan with crowdfunding campaigns – especially during a global pandemic that’s led to supply chain issues. But Morefine is an established company that’s been around for a while and which has brought multiple mini PCs to market over the years. The company seems to be using crowdfunding more as a promotional tool than a fundraising one. http://https://liliputing.com/wp-content/uploads/2022/01/m6_01-780x436.jpg At the heart of the company’s latest mini PC is Intel’s Celeron N5105 chip, which is a 10-watt, 4-core, 4-thread processor with base and burst speeds of 2 GHz and 2.9 GHz, respectively plus Intel UHD graphics with 24 execution units and base / burst speeds of 450 Mhz and 800 MHz. Under the hood the little computer features LPDDR4 memory, M.2 2280 and M.2 2242 slots that can be used for NVMe and SATA storage, respectively, and an Intel AX201 wireless card with support for WiFi 6 and Bluetooth 5.2. Ports include: 1 x HDMI 2.0 1 x USB Type-C (full function) 1 x USB Type-C (power input) 3 x USB 3.1 Gen 2 Type-A 1 x 3.5mm audio 1 x 2.5 Gbps Ethernet Morefine notes that the computer can drive up to two 4K/60Hz displays when using the HDMI and USB-C port at the same time, thanks to USB DisplayPort Alt Mode functionality. And since the other USB-C port which is used for power works with any power supply that supports 12V/2A output, not only can you plug the computer into a wall charger, but you may be able to power the system from some USB power banks, enabling on-the-go usage. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/25/22 at 19:23:49 Several real world comparison test results from real laptops are in now. Intel Alder Lake always pulls double the watts of current that the equivalent AMD does, and then Intel Alder Lake can spike up to 3x and 4x higher depending on the test that is being done. This is "the new Intel Alder Lake normal" and it is within the new Intel specs as were recently announced. Laptop coolers are simply not up to this job, so the Intel Alder Lake laptops quickly cycled down into cool off mode and slow down greatly. The mid-length or longer duration tests are coming up all pro-AMD or sometimes as a near draw, depending on the test's duration. Everywhere you can see this cooler war going on, with AMD not needing nearly as much cooling (roughly half as much) as Intel requires. The corollary to this cooler war is AMD can stay running faster for longer, and can apparently run as fast for longer than Intel Alder Lake can do in the current match up pairs if given a test of some duration. On the shorter laptop tests, Intel rules. A very short test can show a 25-30% Intel win advantage which quickly rolls down into negative numbers as the elapsed time of the test increases. Simple, huh? Clear as mud, huh? ;D ...... Another corollary, Intel recommends that you leave their laptops hooked up to their high wattage chargers at all times during use ...... (if you don't, Intel's batteries tend to go down fast). |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/25/22 at 21:56:38 https://www.tomshardware.com/news/nvidia-arm-acquisition-expected-to-fail-nvidia-loses-dollar125b-deposit-report Nvidia's Arm Acquisition Expected to Fail, Nvidia Would Lose $1.25B Deposit By Mark Tyson published about 16 hours ago At least two insiders have talked to Bloomberg about the $40 billion deal's impending collapse. If the murmurings of Bloomberg's sources are to be believed, we are at a very interesting pivot point in the long saga of Nvidia's attempted ~$70 billion acquisition of Arm. The heavyweight financial journal cites two unnamed sources behind its assertions that Nvidia is quietly preparing to abandon its Arm takeover plans. The deal falling through will undoubtedly be particularly painful for Nvidia because it agreed to pay SoftBank Group $1.25 billion if the Arm acquisition failed. Nvidia pre-paid the penalty when the deal was announced. A 1.2 Billion dollar ouchie for Nvidia for failing to gain regulatory approvals from ANYBODY for this deal. NOBODY wants it to go forward. I wonder if Bennie "Leatherman" Huang's egomania learns anything at all from this lesson about his "I can do whatever I want" fixations. Bennie's blowing off the FCC lawsuit was an act of purest egotism ....... some big fines and more FCC lawsuits to follow. If Benny continues his train of egotistical nonsense the FCC will eventually fine him a series of healthy fines in addition to the 1.2 billion "failure fee" he has already pre-paid to ARM already to start his bid rolling (and that failure fee is already forfeit due to Benny being well past the last final close date having gained ZERO regulatory approvals), this all being compounded by the FCC's pointed lawsuit direction to Benny not two weeks ago telling him to immediately abandon his take/over merger attempt as a failed effort and Benny choosing to willfully not comply with the controlling governmental body. 1/26/22 New Info ....... China regulators just approved the AMD/Xilinx merger, it is a go ASAP |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/26/22 at 22:38:29 https://www.theverge.com/2021/7/26/22594074/intel-acclerated-new-architecture-roadmap-naming-7nm-2025 OK, here is a restatement of where Intel really is vis-a-vi Apple M1 and TSMC's 6nm and 5nm as used by AMD. Intel is functionally at the old Intel 10nm still, but Intel has taken delivery of one each of a two generations back ASML 5nm lithography machines. These are not up and running yet. THIS IS FROM THE SUMMARY AT THE END OF THE ARTICLE Intel’s new names may help the company recontextualize its current and future products more accurately against its competition, but the fact remains that Intel is behind. Even accepting that the Intel 7 is on par with 7nm products from other foundries, those foundries have gone on and are already past their 7nm chips and on to 5nm hardware at this time. Which means that the companies that rely on those external foundries — like Apple, AMD, Nvidia, Qualcomm, and virtually every other major tech company — can still get chips that are more advanced than Intel’s best work. Apple’s superlative M1 Macs, for example, already use 5nm chips from TSMC — and handily outpace Intel’s comparable products. AMD is rumored to be working on 5nm Zen 4 processors for as early as 2022, too, which could offer similarly concerning competition for Intel from its already encroaching competitor. Even with the ambitious, annual cadence for its roadmap, Intel is playing from way behind Apple and the others; Intel doesn’t expect to fully catch up to the rest of the industry until Intel 20A in 2024. And it doesn’t expect to reclaim leadership in the semiconductor business until 2025 with Intel 18A. And all that assumes that Intel doesn’t hit any more delays or manufacturing snags like the ones that held up both its 10nm and 7nm processes (which arguably put the company in its current situation in the first place). It also assumes the world holds totally still for 3 years instead of the rest of the world continuing making its current much more rapid progress than Intel can do by itself. Right now all Intel has done is to rename all of its production nodes and to codify Intel's extreme power draws and the extreme heat production of its current processors, claiming this to be "world beating progress" and the new normal. ::) i.e. pure bullshite ...... Intel's real progress of late has been in building some much better desktop processor coolers and in attempting (and falling) to do the same in laptops. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 01/27/22 at 16:42:15 https://liliputing.com/2022/01/xmg-oasis-is-an-external-liquid-cooling-system-for-the-xmg-neo-15-gaming-laptop.html http://https://liliputing.com/wp-content/uploads/2022/01/xmg-oasis_01-778x500.jpg This rig is what it takes to effectively cool down a major Alder Lake laptop chipset ....... a desktop sized fan and radiator based cooler rig. "Hold on for a sec guys, I have to dribble all of the green stuff out of my laptop's cooling port or else it will make a bloody mess in my book bag again iffen the little snap cap comes off the brass connector again ......." Reality talk now, people are not going to be able to run an Alder Lake processor near its capabilities inside a realistic laptop format. Intel laptops will eventually get compared to similar units from AMD, and the Intel units may be seen as equivalent ....... for a short while until they completely overheat, anyway. Intel's "drive it hard and really heat it up and then lie about it" approach will certainly fail against Apple and against 5nm AMD products because these guys simply don't have the heat it up issues at anywhere near the same level that Intel has. Plus, any realistic tech that Intel can do to mitigate their Intel overheating could also be applied to AMD and Apple laptops, yielding even better results for Intel's competitors in direct task based comparisons. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/05/22 at 05:18:37 https://www.channelnewsasia.com/commentary/intel-chip-semiconductor-shortage-pc-phone-tsmc-qualcomm-2478686 I have been saying Intel has a shaky house of cards as their foundation and that Intel LIES A LOT about their performance, cost, energy usage and a bunch of other things. News organizations see Intel through their own set of filters, and report accordingly. This reporter splits Intel up into design and manufacture groups and then shows how Intel has failed in both areas. Every perspective I get shown lately says Intel is going down while Pat Gelsinger keeps saying that Intel has just hit rock bottom and the big turn around is happening starting tomorrow. https://www.channelnewsasia.com/commentary/intel-chip-semiconductor-shortage-pc-phone-tsmc-qualcomm-2478686 http://https://onecms-res.cloudinary.com/image/upload/s--hXURAa4U--/c_fill%2Cg_auto%2Ch_468%2Cw_830/fl_relative%2Cg_south_east%2Cl_one-cms:core:watermark:ap_data-1%2Cw_0.1/f_auto%2Cq_auto/v1/one-cms/core/malaysia_intel_01111.jpg?itok=rz2a3U14 LAUSANNE, Switzerland: American chip-making giant Intel is a shadow of its former self. Despite the global semiconductor shortage, which has boosted rival chipmakers, Intel is making less money than a year ago with net income down 21 per cent year over year to US$4.6 billion. Unfortunately, this is an ongoing trend. ::) Intel just now took delivery on two different ASML machines (one machine per shipment of slightly different 2-3 year old technology) that Intel does not know how to design processors to be run on, nor has any practical experience in running these ASML machines. None of Intel's current stuff runs well on anything other than Intel's decrepit in house 10 and 14nm process nodes. Intel is faltering in both design and in manufacturing and even news organizations are reporting this now as "hard news". 80 billion dollars of one shot Biden money will not fix Intel. Pat Gelsinger is too old to eventually rejuvenate Intel as by Intel's own rules he will be hitting mandatory retirement in 3-4 years. Intel has only served to spin up the TSMC and AMD cooperative progress approach. APPLE still funds TSMC's lithographic progress with large blasts of "state of the art only money" on a yearly basis. Nether Apple nor Samsung nor AMD are interested in using Intel as a foundry. Look to Global Foundries as an example of what Intel will become once the current "chip shortage" ends starting next year. Intel is seeing this happen at 28nm and 32nm right now as better more modern nodes come on line at TSMC and at Samsung. Intel utilization rates on these nodes are falling monthly. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/07/22 at 20:39:10 https://liliputing.com/2022/02/intel-invests-1-billion-in-third-party-chip-development-joins-risc-v-international.html http://https://liliputing.com/wp-content/uploads/2022/02/ifs-715x500.jpg Note how egotistical this slide actually is, it shows the Intel first mentality very clearly. It shows RISC as a minor small addendum to a huge Intel mainframe combo chipset. Intel invests $1 billion in third-party chip development, joins RISC-V International by BRAD LINDER Intel invests $1 billion in third-party chip development, joins RISC-V International at the at the Premier Membership level Intel is probably best known as a company that designs and produces its own chips. But the company has also been investing heavily in its Intel Foundry Services (IFS) recently in hopes of taking on rivals like TSMC and Samsung by manufacturing chips for other companies. Now Intel has announced a new $1 billion fund to help spur development of chips that may make use of the companies foundry services… including x86, ARM, and RISC-V processors. And it’s placing a pretty heavy emphasis on that last one. RISC-V is an open instruction set architecture that allows chip makers to use the core technologies without paying any royalties. The architecture’s been around for more than a decade, but which has really begun to pick up steam in recent years, although the fastest RISC-V chips today still trail far behind the latest Intel or ARM processors when it comes to performance (although this kind of investment could eventually close bridge the gap). Intel says IFS is the only major global foundry that’s prepared to manufacture x86, ARM, and RISC-V chips and that it’s able to produce RISC-V processors for third-party partners as well as “chiplets” which are basically custom-purpose chip blocks that can be put together using 3D stacking to deliver a system-on-a-package with multiple blocks that each serve a specific purpose. In other words, if I’m reading Intel’s announcement right, you could see packages that combine an x86 or ARM processor and a RISC-V co-processor. As part of today’s announcement, Intel says it’s joined RISC-V International and announced partnerships with leading RISC-V companies including SiFive, Andes Technology, Esperanto Technologies, and Ventana Micro Systems. While Intel has been practically synonymous with its x86 chips for decades, this isn’t the first time we’ve heard about the company’s interest in RISC-V. In fact, last year there were rumors that the company was interested in acquiring SiFive for $2 billion, but that deal eventually fell apart. Coloring inside the lines for once (and doing an affordable small smart thing since their older production lines are going empty) Intel has shown they want to be one of the main foundry sources for this up and coming RISC-V processor waves. But don't consider Intel to be totally benevolent in doing this brand new billion dollar fund of theirs (which is actually their entry fee dollar amount for entry into RISC International at the premier membership level). What Intel is actually planning on doing is planning on spending the billion dollars to build up their own Intel tool kit and create a good "little guy communications systems" to properly handle the budding new RISC-V business if they ever get any of it. Intel is notably unfriendly for small guys to deal with right now, and that has to change fast or Intel will lose this new change wave just like they have done to so many other recent change waves. Intel's older processor production lines ARE NOT DESIGN OR BUILD COMPATIBLE to ARM or to TSMC's existing design standards that the small guys all use and you must re-do your designs to Intel's rule book to be able to use Intel as a foundry. This redesign for non-compatible production processes locks you into Intel which isn't always such a good idea. Remember, Intel always gives its own production needs priority over outside customers ...... These are the same rocks that scuttled the past 3 attempts for Intel to do foundry work for other people in the last 5 years. Unless Intel drops their foundry prices CONSIDERABLY right now and gets a whole lot easier to deal with like right now fast, then TSMC will own this budding business anyway inside the next 2 years. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/08/22 at 11:01:28 https://liliputing.com/2022/02/nvidia-removes-arm-from-its-shopping-cart-arm-will-go-public-instead.html NVIDIA removes Arm from its shopping cart, Arm will have to go public instead by BRAD LINDER 02/08/2022 NVIDIA’s plan to acquire chip designer Arm Holdings is dead. The companies announced the deal a year and a half ago, but since then it’s come under heavy scrutiny from antitrust regulators around the globe, and now NVIDIA and Arm’s parent company Softbank say they’ve terminated the agreement. At the same time, Softbank is unveiling new plans for Arm. It’s appointed a new CEO for Arm and announced plans to take the company public by March 31, 2023. Arm doesn’t make its own processors, but the company designs chip architecture that’s licensed by makers of virtually every smartphone on the market today as well as many other devices including consumer electronics, servers, and even recent laptop and desktop computers like Apple’s latest Macs with Apple Silicon chips (which are based on Arm technology). Not surprisingly, many chip makers were immediately skeptical when a rival chip maker (NVIDIA) announced it wanted to buy Arm, but would not give its own chips preferential treatment in any way. And it seems like antitrust regulators in the US, Europe, China, and other regions were also concerned. With the chances that the deal would survive that level of scrutiny looking slim to none, Softbank and NVIDIA decided to go another route. NVIDIA will continue to hold a 20-year license for Arm designs and Softbank will get to hold onto $1.25 billion that was pre-paid by NVIDIA. Japanese company Softbank acquired the UK-based Arm Holdings in 2016 for $32 billion with an announcement that it would pump money into the company to grow Arm’s global business. According to Reuters, Softbank would have received around $12 billion in cash and up to $50 billion worth of NVIDIA stock if the sale had gone through as proposed… but with the deal falling apart, Softbank will have to look for other ways to raise money. And it looks like the company is hoping a public offering will do that. Arm’s new CEO who will help lead that initiative is Rene Haas, who’s been working in the chip industry for more than three decades, including 7 years as a vice president at NVIDIA and, most recently, 8 years on Arm’s executive team. ARM brutally takes it in the shorts on this one as Softbank is keeping NVIDIA"s 1.25 Billion dollars and Softbank is granting NVIDIA a 20-year license for Arm designs while Softbank is busy fleeing the scene. :o ARM is being forced to go public and will lose a LOT of value in that control exchange. Look to see the reorganized new ARM have to do something about this totally inequitable deal that is being forced on it by Softbank (and by NVIDIA) once it is finally free of Softbank's control. Can you say "reorganization" by splitting itself up with NVIDIA holding a 20 year license from the weakest remaining part? Also, look to see RISC-V make a big big uptick in the marketplace as ARM is being gravely wounded by this deal. The same regulators that turned down the NVIDIA acquisition may well pass judgement on this Softbank NVIDIA screwjob and actually act to protect ARM. Mebbe ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/10/22 at 11:43:15 https://liliputing.com/2022/02/amds-acquisition-of-xilinx-set-to-close-within-days.html NVIDIA’s high-profile acquisition of chip designer ARM might have fallen apart this week. But consolidation continues to be a trend in the semiconductor space: AMD has just announced that it’s acquisition of Xilinx is set to close “on or about February 14, 2022.” http://https://liliputing.com/wp-content/uploads/2022/02/xilinx-ultrascale.jpg AMD had announced in October, 2020 that it planned to buy Xilinx for $35 billion, and nearly a year and a half later the company says it’s received all the necessary approvals from government regulators, which will allow the deal to close soon. AMD has said that its purchase of Xilinx will help the company expand its offerings for data centers, but FPGAs can also be used in 5G networking gear, automotive, industrial, medical, and security applications, among other things. Lisa Su at AMD has grace and tact, and she isn't going run around rubbing any salt in anybody's wounds right now. HOWEVER, she is going to chair one combined company as of this upcoming Monday, one company that is going to get a big up tick in the stock market when the deal closes on this upcoming Monday. Look to see some serious Xylinx based processing accelerators going into AMD mainframe chipsets, with some smaller versions likely to turn up in Threadripper soon enough and into the 9700 series whenever it gets announced later this year. Xylinx invented the AI accelerator and the FPGA and they are the best in that sort of business, bar none ....... Intel has been counting on their Altera tech to boost Intel's processing power for a quite a while now. Look to see the same sort of thing coming from AMD as they are going forward into their future lithography generations. =================================================== OK, the stage is set. Intel has rolled out all of the rest of their Alder Lake chipsets and Intel has heated up all the computer rooms of all the folks who believe in Intel's copious BS testing tricks and all of Intel's quite real power quadrupling SUPER hyper clocking & high energy draw performance boosting effects. Many folks don't want a noisy room heater for a computer, and they are waiting to see what AMD's next generation of processors can do to separate the room heating from the computing activities. HOWEVER, Intel has shifted the playing field some and to be a winner now-a-days means that AMD is going to have to push out some larger computing throughput levels too. ;) ...... betcha AMD can do it without becoming a noisy power sucking room heater. 2/11/2022 AMD news AMD has now admitted to rolling their 5nm generation forward a quarter or two, not specifically saying it is in response to Intel's unexpectedly good showing with Alder Lake, but AMD is rolling it forward just the same. Intel is now preparing their rebuttal to AMD's 5nm for early next year and is hoping to not have to deal with a Xylinx amplified consumer AMD processor at this particular point in time. Pat Geslinger's house of cards is shaking and moving around right smartly right now ........ Gelsinger is putting pressure on AMD and that is causing AMD to make more rapid progress motions. We like competition that leads to real amounts of measurable progress --- and now it is AMD's turn to waggle their butts and SIMPLY AMAZE us. =================================================== Apple and TSMC move forward on 3nm (source Apple news) https://9to5mac.com/2021/12/02/3nm-iphone-ipad-mac-chips-tsmc/ TSMC has kicked off pilot production of chips built using N3 (namely 3nm process technology) at its Fab 18 in southern Taiwan, and will move the process to volume production by the fourth quarter of 2022, according to industry sources. Why is this important? As Apple moves off 5nm AMD takes their place. Next, both Xilinx and AMD both have large already paid for contract production runs scheduled on TSMC 3nm. TSMC 3nm is actually booked up tight for the next several years ........ ...... how in the world Intel says they are moving bulk production to TSMC 3nm in this time period is beyond me. All Intel has prepaid & booked is a few graphics processors at 3nm and that is all they got showing on 3nm. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/14/22 at 02:51:21 https://www.bloomberg.com/news/articles/2022-02-09/intel-loses-market-share-in-server-chips-but-makes-gains-in-pcs Intel Corp., working to regain its edge in processor technology, lost more ground to rival Advanced Micro Devices Inc. in the lucrative server-chip industry during the fourth quarter but took back some market share in personal computers. Altogether, AMD reached a record high of 25.6% share in processors in the period, including custom chips for game consoles and semiconductors for the so-called Internet of Things, according to Mercury Research. That surpassed a previous high of 25.3% set in 2006. Intel chips still account for most of the market, but the company doesn’t have the technological supremacy that it once held. That’s led customers to seek alternatives, whether it’s AMD processors or chips of their own design. Intel Chief Executive Officer Pat Gelsinger has promised to reverse gains made by AMD with new products that will outperform its longtime rival. But the PC chip market has been booming, which is good news for both Intel and AMD. Revenue from their types of chips surged 11% to $74 billion in 2021, Mercury found. Chips made using Arm Ltd.’s technology also grew, bringing the total industry to 500 million units in 2021. AMD enjoyed strong demand for game consoles made by Sony Corp. and Microsoft Corp., a market it dominates. Its share of the server segment rose to 10.7% in the fourth quarter, up 3.6 percentage points from a year earlier. Intel once controlled more than 99% of the industry. In desktop PCs, Intel’s share of the market rose to 83.8%. It was 78.4% in laptops. Arm-based chips, meanwhile, were once a tiny niche of the PC processor market. But Apple Inc. switched its Mac line over to that type of semiconductor, giving the category a boost. The market share of such chips nearly tripled from a year earlier to 9.5% in the fourth quarter. OK, in a growing booming (11% net growth) computing industry Apple shows the largest provable growth numbers (9.5%) with their ARM based chips. AMD is next biggest grower having taken 10% of server segment (ie. almost all the server industry growth went to AMD this year). All of the new console business went to AMD. AMD lost out most on the industry growth in PC and Laptop areas with Intel taking the most (but not all) of that growth. Intel grew better in the PC and Laptop areas (taking back some share from AMD but losing a significant chunk to Apple). Intel lost more than the industry growth, making Intel a net shrinking company again last year. So, Intel lost more ground than Intel gained. AMD gained slightly overall, winning some and losing some. Apple cleaned up the most with significant market share gains with no downside losses. BIG PICTURE TIME AGAIN Intel remains 83% of the total installed computing market share and you could say that AMD and Apple are growing simply by swiping the crumbs off the dining room table compared to Intel's gigantic historical presence. But the fact is this current sales net loss trend is continuing with Intel not growing while everyone else does significant growth ....... Intel's losses are approximately 103% all of its potential industry growth numbers --- INTEL IS SIMPLY NOT GROWING LIKE ALL THE OTHERS ARE DOING. I slam Intel for pushing lies and making room heaters instead of computing devices, but the hard fact is that Intel HAD to do these BS tricks or Intel would have shed even more market share, shed it like spring lamb's wool during this stressful time. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/14/22 at 04:31:01 Cheap easy prediction time ........ ARM chip designs from Mediatek will join Apple, Samsung and Qualcomm and Amazon in taking market share from the low end (Chromebook side) of the x86 market in 2022 (with Amazon using their own ARM processors in their Amazon Web Service servers on the top end. Mediatek's growing presence in Chromebooks will begin lapping over into low end Windows Laptops later on this year or next year. Mediatek will grow and monetize even more on the sliver of growth that Samsung and Qualcomm had carved out with the very first (and these being the most expensive) ARM PC chipsets. A reinvigorated ARM Holdings that has been freed from Softbank control will help this market share grow even faster by making even stronger laptop chipset designs available that any of the phone boys can make once they license the design. The first widely accepted free to build RISC-V products will likely come out this year, starting up that source of inroads on the x86 market share decline. AMD not wasting any more development dollars or production wafer counts on the lowest end x86 processor sets as these are "already gone" in AMD's eyes. This position doesn't have much to do really with Intel or what Intel is doing per se --- every one of the 3-4 current ARM suppliers is busy slicing up on that particular part of the pie and the odds of an AMD x86 processor ever taking back all of that finely re-sliced piece of pie is very slim indeed. I look for Mediatek or Rockchip (possibly in concert with Google) to come in from the side with a winning and very compelling low end laptop product on the lowest end of things and make moves to own that market segment moving forward. Or mebbe Amazon with a Mediatek processor Fire Chromebook Tablet with a dedicated fold over holder/keyboard affair ........ the very lowest part of the laptop spectrum is rife with cost saving innovations like this. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/17/22 at 04:43:07 News from 2/17/22 Tom's Hardware and at least 3 other sites along with Motley Fool report that post merger AMD is a larger stock based asset company than Intel is at this point in time. Don't get too excited, that will subside some when the AMD buying frenzy is over and the stock price drops some. Intel is saying they will put x86, RISC-5, ARM, MIPS all on the same chipset for you to your design. Nice words, Mr. Geslinger, now let's see Intel go build a good first lot of foundry chips for somebody, anybody as Intel has failed to do so 3 times in the historical past and Intel is still basically Intel to the best of my knowledge. TSMC could do the same sort of mixed up stuff, but nobody really wants to do that in the real world. ================================================== Intel breaks 300+ watts power draw on Intel's latest I9 room heaters. http://https://cdn.mos.cms.futurecdn.net/Gp2zGqUYDcFcCUfywW5S7n-970-80.png.we https://www.tomshardware.com/reviews/intel-core-i9-10900k-cpu-review/2 Look at the graph and think for a moment. Intel overheats and quickly slows down to the point it is about the same throughput as AMD's current processors. AMD really is half to a third of the watts draw that Intel needs for their current trickery. AMD is getting ready to drop down to a real 5nm process at which Apple's fans never even came on during use. AMD is going to incorporate Xilinx AI inside their chiplet based processors within the next year. AMD is also incorporating improvements in cooling technology that is absolutely required by Intel, but AMD will get processing boosts out of it that Intel will never see. All Intel has in their corner is big watt hyper-clocking BS and even bigger marketing lies ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/18/22 at 09:41:41 https://liliputing.com/2022/02/intel-outlines-roadmap-for-raptor-lake-meteor-lake-arrow-lake-and-lunar-lake-chips.html http://https://liliputing.com/wp-content/uploads/2022/02/intel-roadmap-2024-780x462.jpg Intel spells out their marketing lies for the next 3 years. Read the thing and look at the pretty graphics ---- yes that next two generations covered in the first slide is really just the same Intel 10nm all over again as Intel simply hasn't got their 7nm process (Intel 4) working right so Intel can roll over into it. New BS names, even higher current draws and even bigger heat levels produced ---- but still Intel 10nm. Beyond that, in the short term it all depends on if Intel can get TSMC to run whole lots of Intel chips in bulk at 5nm and at 3nm for them or not. Intel 4 (alias the old Intel 7nm) still has ongoing overheating and performance issues and very low yields so don't be counting on Intel 4 roll in to save the day any time soon. Intel NEEDS TSMC 5nm and 3nm ....... very badly. Look for the purple blocks that indicate TSMC is hopefully running the stuff for Intel. Now, if TSMC decides NOT to support a competitor fab named Intel, this will never happen. And Intel immediately fails ...... Another variable is the brand new 5nm lithography scanner equipment Intel just bought from ASML -- can they actually make up an Intel design that actually runs on either of these two slightly different 5nm processes? So far, no luck with that either. ================================================== The press attending the Intel open question segment of the big event yesterday was full of harsh hardball questions for Pat Gelsinger about how Intel was going to ramp up their 10nm and 4nm production levels with so few pieces of equipment that can actually run those processes. One guy even asked Pat how he could explain the ongoing loss of 35% of Intel's old supremacy with 25% being lost in just this last year alone. Pat made noises about turning 10 years worth of bad management completely around next year, but he then got asked pointedly about stopping the bleeding this year and he hummed and hawed a bit over that one right along with the next few questions about Intel not dealing with their processors over heating and the now quite extreme watts of power draw (currently up to 250 watts, or ~200 watts~ more than AMD requires to do the same job). I am not the only voice out there that thinks that Intel is still rapidly failing on several fronts and Intel is simply NOT getting significantly better nearly fast enough. Both computer enthusiasts and stock holders are no longer deluded by Intel's pretty slide shows and positive pitch presentations of the newest Intel room heaters. Pat Gelsinger's house of cards is indeed very shaky. The progression of the hard numbers say that Intel is still busy failing in the marketplace as we watch ....... Intel just took another 10% decline in their stock price while AMD went up 6% in the same time frame. This pushed Intel's "net capitalization" stock valuation to be less than AMD/Xilinx total net value for the first time ever. The EU environmental wackos are now even talking about new regulations limiting the very recent explosive growth in excess "waste computer watts" draw as a "growing environmental concern". Apple and AMD don't have this concern, only Intel runs their stuff as impromptu room heaters. Apple at 5nm didn't even have their fans come on during testing, and at 3nm this is even more unlikely. x86 with 5nm AMD, yes the fans will run ---- but only lower wattage small fans. Within the next 6 months we will see early test results from both Apple at 3nm and AMD at 5nm so then we will KNOW how badly disadvantaged Intel truly remains. AMD's rumor mill is talking about adding core counts to the main PC (and to the higher processor products as well) again due to having lots of socket and die space available due to much smaller and higher density TSMC 5nm lithography. The sources mention adding a "boutique chiplet" with Xylinx AI functionality to the AMD performance mix along with full ray tracing gaming level graphics built into all AMD processors. We will believe it all when we see it, of course. AMD plans to raise their rated current draw to 180 watts or thereabouts, as the extra cores, AI and better graphics will suck some extra power. You still come out ahead on watts used in total because you won't need a dedicated separate 250-300 watt graphics card or to power all those extra sticks of RAM systems memory. Plus, AMD built in on chip Smart Access Memory will be available to all AMD functions, and at 128 gig of Smart Access Memory this is a considerable pool of fast access GPU grade memory to support gaming graphics and some REALLY REALLY HUGE spreadsheets when doing real work type work. ::) |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/20/22 at 07:39:18 In case you think I am all alone in my take on shaky Intel, here is Motley Fool ringing in on the subject. https://www.fool.com/investing/2021/12/13/4-red-flags-for-intels-future/ 4 Red Flags for Intel's Future By Leo Sun - Dec 13, 2021 at 7:00AM KEY POINTS Intel is trying to catch up to TSMC in the process race. But the company's plan relies too heavily on government subsidies. It might need to cut its dividend to gain more financial support. The chipmaker's stock is cheap for obvious reasons. Intel ( INTC -5.32% ) might initially seem like an attractive investment for value-oriented income investors. The chipmaker's stock trades at just 10 times forward earnings and pays a forward dividend yield of 2.7%. Intel's insider sentiment has also improved since Pat Gelsinger took over as its new CEO in February. Over the past 12 months, Intel's insiders bought nearly twice as many shares as they sold. But during that period, Intel's stock rose just 5% as the S&P 500 advanced nearly 30%. AMD's ( AMD 1.30% ) stock skyrocketed more than 50%. Intel's low valuation and high yield might limit its downside potential, but four red flags could also prevent it from outperforming the market. 1. Its data center share losses to AMD Between the fourth quarters of 2018 and 2021, Intel's share of the global x86 CPU market fell from 77.1% to 62.1%, according to PassMark Software. AMD's share rose from 22.9% to 37.7%. Intel's losses in the desktop and laptop markets were already disappointing, but its losses in the data center market -- which it traditionally dominated with its high-end Xeon CPUs -- were the most surprising. During that period, Intel's share of the server market fell from 98.4% to 91.9%. AMD's share rose from 1.6% to 8.1%, with most of those gains occurring earlier this year. That pressure clearly indicates that AMD's EPYC chips are pulling big data center customers away from Intel's pricier Xeons. If Intel can't stop that bleeding, its data center group (DCG), which generated 34% of its revenue last quarter, could be in serious trouble. 2. Its ambitious plans to overtake TSMC and Samsung Intel ceded the CPU market to AMD because it suffered from delays and chip shortages while falling behind TSMC ( TSM -0.38% ) and Samsung in the "process race" to manufacture smaller and more advanced chips. Intel manufactures its own chips, while AMD outsources the production of its highest-end chips to TSMC. So when Intel's own foundry failed to keep pace with TSMC's smaller nodes, it also lost its technological lead to AMD. Before Gelsinger returned to Intel, his predecessor Bob Swan had flirted with the idea of becoming a "fabless" chipmaker like AMD. However, Gelsinger quickly scrapped that idea and doubled down on expanding Intel's manufacturing capabilities by investing billions of dollars into building new plants in the U.S. and Europe. Intel is also opening up its plants to third-party fabless chipmakers to compete against TSMC and Samsung. Gelsinger believes Intel will catch up to TSMC and Samsung in the process race by 2024 then reclaim the lead by 2025. That bold claim raised a lot of eyebrows since TSMC already plans to spend about $100 billion over the next three years to expand its capacity and maintain its lead. Therefore, Intel will likely need to outspend TSMC to reclaim the lead in just four years, but it only allocated $18 billion to $19 billion to capital expenditure (CapEx) this fiscal year. 3. Its dependence on big government subsidies Intel doesn't have enough cash to regain the process lead on its own, so it's asking the U.S. and Europe for large subsidies. In the U.S., Intel is pushing for the approval of the CHIPS act, which would grant $52 billion in subsidies to domestic chipmakers for the production of new stateside plants. Intel is also asking the European Commission to fund its development of new plants with nearly $10 billion in subsidies. Intel claims that its plants will enable fabless chipmakers to reduce their dependence on TSMC, Samsung, and other Asian foundries. Gelsinger also argues that Chinese threats make Taiwan -- which is home to TSMC and its smaller rival UMC -- a more "unstable" region for chip production than the U.S. Unfortunately, Intel's plan has a glaring flaw: The U.S. government views Taiwan as a crucial geopolitical ally, and it already granted TSMC subsidies for the construction of a new $12 billion plant in Arizona earlier this year. TSMC has also been urging Congress to include foreign chipmakers in the CHIPS act. If that happens, Intel's entire subsidy-funded expansion plan could fall apart. 4. It needs to suspend its dividends Intel has already taken bold steps to raise more cash. It's in the process of selling its NAND memory chip business to SK Hynix for $9 billion, and it plans to spin off its automotive chip unit Mobileye in an initial public offering (IPO) to raise more cash. It also significantly reduced its buybacks this year. But, Intel could free up significantly more cash by suspending its dividend, which used up $5.6 billion of its free cash flow (FCF) in 2020 and another $4.2 billion of its FCF in the first nine months of 2021. It doesn't make any sense for Intel to pay out billions of dollars in dividends while pleading for government subsidies. Intel could eventually realize that it's a bad look and suspend its dividend, but its income investors will then flee. A rocky road ahead Intel's stock looks cheap, but these four red flags justify that discount. It's still struggling to keep pace with AMD, it's heavily dependent on future government subsidies, and it might need to suspend its dividend in the near future. Investors should stick with better-run chipmakers until more green shoots appear Fifth red flag: Being totally dependent on TSMC for mid-term survival Intel must be aware that China potentially invading Taiwan is still feared or else TSMC could simply proactively act in their own best interest ---- this means that Intel's mid term functionally risks taking a potentially hard blow sometimes year after next. As a self-announced competitor who firmly intends to take chunks of TSMC existing business Intel may find itself paying "non-preferred" rates and tougher terms payable in advance to run their chiplets at TSMC. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/21/22 at 21:09:26 https://youtu.be/_jX-hKvUQDU Lisa Su brings out the 6nm APUs with AMD gaming graphics built into the onboard video system. This entire generation will be short lived as AMD is rolling the 7000 series forward by half a year, putting the 5nm 7000 series out in the second half of this year. Why? Intel had uncorked a surprise with their Alder Lake BIG/little chipsets and AMD is moving their stuff forward half a year early to act to squelch Intel's attempts at taking back a lot of the PC market share Intel had lost to AMD over the last few years. AMD has firm 6nm commitments to both producers and customers that they will honor, but AMD will not be slow about producing, installing and then replacing the 6nm products that they had promised for this spring. Yes, starting second half of this year 5nm Ryzen 7000 rolls in with even more profound improvements, all of which take place with reasonable power draw watts compared to the Intel room heaters we have been seeing from Alder Lake. So, AMD 6000 6nm will come and go in the space of six months ....... Competition is a good thing ---- and AMD has got some serious product advancements in Ryzen 7000 to give back to Intel to see how well they can respond to it. I liken it to an old swashbuckling sword fight going up a broad set of stairs. Back and forth they go, going up step by step getting better and better all the time. Now, should you buy a Ryzen 6000 6nm AMD processor? No, hold your breath and buy the 5nm Ryzen 7000 product that will be out very soon, and remember the 5nm stuff is going to be far better than what AMD is announcing today. ==================================================== Looking forward to late next year, AMD 8000 will be optimized for TSMC's 4nm, an improved version of 5nm run off the same equipment, instead of dropping down to TSMC 3nm which was to be run off of new lithography equipment. TSMC is having serious yield issues with the 3nm process that makes Apple possibly the only user of TSMC 3nm in the next two years due to TSMC's simple inability to make enough good processors to be shipped for Apple. AMD is blessed with their chiplet based designs, as they can use a variety of chiplet sizes to make up their state of the art assembled chipsets. AMD is inherently flexible, in other words, and can work with whatever chiplets happen to yield out the best. Monolithic designs lack this flexibility. Intel will surely suffer high scrap rates if their large monstrosity of a chipset has any big yield issues on TSMC 3nm. Chart showing percent prepaid orders for 7/6nm, 5/4nm and 3nm CPU and GPU manufacturing services from TSMC http://https://www.club386.com/wp-content/uploads/2021/12/tsmc-customer-chart.png If Intel wants more from TSMC it must prepay for it LIKE RIGHT NOW. Intel, lacking published TSMC orders other than for a few graphics processors, Intel really lacks credibility right now in the sub Intel 4 realm. Intel 4 (alias Intel 7nm) is very shaky as it simply doesn't work well or have any sort of yield, but that situation has been known for a while now. TSMC is looking to expand in TSMC locations not in Taiwan, as the Chinese are planning their takeover attacks as we speak. Joe Biden is toothless as the Russians have so recently proven to the world. Biden has stated many times he will not use nuclear anything on anybody, which now invites raw Russia and China aggression using their nuclear arsenal threat. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/23/22 at 05:28:39 https://wccftech.com/samsung-investigating-fraudulent-4nm-advanced-chip-technology-yields-report/amp/ Samsung Investigating Fraudulent 4nm Advanced Chip Technology Yields By Ramish Zafar / Feb 22, 2022 10:51 EST South Korean technology giant Samsung Electronics is purportedly investigating fraud at its chip manufacturing division Samsung Foundry. Samsung Foundry is only one of the two contract chip manufacturers in the world that is capable of producing semiconductors on the advanced 5-nanometer (nm) chip process. The efficiency of such processes is determined by what is referred to as a 'yield' in the semiconductor industry. A fresh report circulating in the Korean press is now claiming that executives at Samsung Foundry might have fabricated the data for 5nm and 4nm yields, in an effort to misplace the funds allocated by the company to improve chip manufacturing efficiency. Samsung Investigates Chip Yields As Products Fail To Meet Order Volume Speculates Report The report, which surfaced earlier today in Korea, comes courtesy of Infostockdaily, and it mentions the yields for Samsung's latest chip manufacturing technologies, which are the 5nm and the 4nm process nodes. The yield of a chip process refers to the number of chips in a wafer that meet quality testing standards, and it is a key parameter that is evaluated by foundries through early production runs before full scale manufacturing kicks off. According to the Korean publication, officials at Samsung Electronics are conducting an investigation on the whereabouts of the funds that had been allocated to Samsung Foundry for improving its chip manufacturing process yields. This is to determine whether the figures for the yields of the 5nm and 4nm nodes were falsified in order to create an impression that all is well at Samsung's chipmaking arm when reality would suggest otherwise. Samsung and the Taiwan Semiconductor Manufacturing Company (TSMC) are the only two in the world that are selling chip products to customers under the 5nm branding. The pair are currently locked in a race to be the first to deliver 3nm products to the market. This internal investigation results from Qualcomm hitting only 35% yield rates on their last batch of Samsung chipsets. TSMC and Apple are having similar results on the latest runs of Apple chipsets and are struggling to improve them on 3nm. ASML is under the gun to prove to everybody that their latest multi-billion dollar scanners actually work as claimed. Blame is currently flowing towards the wafer material suppliers who are suspected of having supplied sub-standard raw materials as the wafer suppliers have also been an issue of late to Apple/TSMC as well. "Wafer perfection levels" are under strong scrutiny right now as we go down in lithography levels to 3nm. What worked OK at 5nm simply isn't good enough at 3nm and below. Both TSMC and Samsung have run trial lots with good acceptance rates within the last year, but finding you cannot trust their suppliers (yes, you Samsung) not to fudge their internal development data -- which means it is rather stupid to continue this headlong race to get down to newest lithography levels ---- and you can now see why AMD takes the very deliberate downsizing pace that they take. AMD does not want to lead the pack, it goes for second or third when getting into new lithography levels and then only if yields are over a certain rather high percentage. Apple is feeling this yield loss pinch right now as well --- as will Intel when they jump in with both feet at 3nm starting next year. |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 02/25/22 at 11:29:02 RUMOR TIME Intel foresees TSMC being sidelined by a Chinese military take over of Taiwan. When this happens, Intel will ask for and will quickly get 100 billion dollars from Biden to build an AMERICAN BASED MODERN chip making capability of some size. This will be a military sort of program and it will move quickly. ASML has already been told by their own Dutch government to expand their scanner building capability to support such a rapid construction of new fab sites, as the Dutch government also sees this potentially happening soon. NATO as a whole is concerned over chip supply ...... Folks theorize that the Ohio and the Arizona complexes will be expanded to their second stage size and that their original plans included this expansion scenario. Stage 1 was for well known high yield processes, stage 2 will run towards current state of the art processes. Buckle up boys, it is going to be a wild and crazy ride IF THIS HAPPENS ...... |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/06/22 at 03:52:06 https://www.tomshardware.com/reviews/cpu-hierarchy,4312.html We are getting ready to kick into a brand new wave of Ryzen 7000 5nm AMD progress, so it is important to show the status quo with all of the Intel room heaters clustered up at the top while they are pulling double to 4x more energy to do the same jobs as the AMD units just under them. Remember, when a current Intel Alder Lake chipset overheats it drops its processing speed A LOT just to keep from burning up ........ it quickly becomes "less throughput" than the AMD units listed just underneath it that just kept chugging away at a non-overheating level. But, when recording big industry wide change levels you have to know where you started from to have it all make sense. http://https://cdn.mos.cms.futurecdn.net/eWu8GfYxjKKNwbiE6MyacA-970-80.png.we |
|
Title: Re: AMD & others --- post AMD dominance in 202 Post by Oldfeller--FSO on 03/06/22 at 04:49:38 DEEP BREATH TIME ........ Intel is going to bet the farm on TSMC 3nm super fin very soon. If it works out, Intel becomes undisputed Industry Leader again. Intel's backup plan is to become its own TSMC here in America if China invades Taiwan. This will be very very very messy if it happens. Eventually, when it all works itself out ...... once again Intel becomes undisputed Industry Leader again. If Intel stubs their toe trying to build up Intel FAB as an industry leader and China does not invade, then AMD/TSMC will continue the competition to be the industry leader. ::) Odd isn't it, that AMD only wins if Intel screws the pooch again big time which is exactly what happened in the last 10 years with Intel. |
|
Title: Intel dominance restarts in 2022 ?? Post by Oldfeller--FSO on 03/08/22 at 10:52:12 Notice the thread title change ----- Intel dominance supposedly re-starts in 2022 Yep, great big power draws and all ---- Intel sez it rules the PC world all over again. We are seeing Intel bet the farm on 3nm TSMC starting late this year and rolling over into 2023. Being clear, Intel is competing against AMD using every dirty trick in the Intel playbook and while Lisa Su plays nice (or tries to) she is beginning to lose ground to Chipzilla on several fronts. Joe and Rita Sixpack simply don't care if Intel runs 4 to 6 times AMD's rated power draw to create a 25% throughput advantage in the PC realm. "See, it's better ---- says so right on the box now don't it?" Nobody in mainframe or workstation will buy such power sucking products right now, but if Intel moves down to 3nm they simply won't have that high current draw problem any more. Intel is shifting over to AMD style chiplets too, so that is a cost advantage Intel will glom on to as well. SO, although I personally like AMD better, Intel looks like it is setting up to take over the PC industry again. ::) sigh ...... |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 03/13/22 at 17:10:10 http://https://cdn.wccftech.com/wp-content/uploads/2018/01/intel-spectre.jpg https://www.tomshardware.com/news/amd-cpus-see-less-than-10-performance-drop-from-revised-spectre-v2-mitigations Spectre-v2 isn't going anywhere soon. VUSec, a group of researchers from Vrije Universiteit Amsterdam, recently discovered Branch History Injection (BHI), a byproduct of Spectre-v2 that affects Intel and Arm processors. According to Phoronix's latest report, while Intel has taken a performance hit up to 35%, AMD's processors got off easy at less than a 10% hit. Yep, we have had 4 days of frantic back and forth over Spectre version 2 issues. Here is my nutshell on this mess. THERE IS A SPECTRE VERSION 2 ATTACK possible on both Intel and AMD that simply steps around all the existing hardware mitigations developed in the last 2 years. Taking it all back to ground zero on fixing this mess, it is a brand new mess that is even worse than before. Two very recent Software Mitigations have been written by the Linux boys for the new Spectre v2 attacks. Two main mitigation variations have been written, with only one being seen as effective enough to go forward into widespread distribution. Intel is simply eat up with the stuff and is seeing potential throughput hits of 20-35% just to do a software based mitigation that Intel couldn't even be bothered to write. Intel can't fix their own problems and is acting all denial paralyzed at the moment. Intel is currently spending all their time in smear campaigns against AMD products instead of trying to fix their own stuff. Intel is looking at 20-35% hits in throughput on common processor tasks. AMD is looking at 5-10% throughput hits because of AMD architecture differences. AMD has options to potentially cut down their mitigation hits to less than 5% by using on chip hardware mitigations that is built into the newest AMD processors. This Spectre stuff totally wipes away Intel's high watts draw "processing advantage" and puts Intel Alder Lake functionally firmly back in #2 position across the board. Intel must build their own hardware mitigations into their new silicon to be competitive to AMD. Intel has now dropped back into their Intel default lying mode as that is all they have right now ........ BS, bad data and bigger lies. AMD is going to try to address this mess in hardware with their newest processors as AMD has a re-programmable on board security chip built into their new hardware that can actually do this job --- while Intel has NOTHING effective that they can do at the moment other than the software mitigations that have been written by somebody else. All of this very bad situation will change and evolve very quickly. I look to see Intel try to ignore it all since they cannot fix it without a huge throughput hit. I look to see MS adopt the best of the Linux boys software mitigations and put it out internal to Win 11 very soon. I look to see the processor ranking shift some when this happens as Intel cannot ignore the performance hit to their processors any longer. =================================================== two days later Is your old machine affected by: 1) the Spectre / Meltdown version 2 new illnesses and their software mitigations --- yes it is. 2) AMD is relatively less affected by the two illnesses, losing less than 10% functionally by existing software mitigations and less than 5% by using the new proposed hardware mitigations that AMD is pursuing. (only new machines need apply for this hardware fix) 3) Intel is being hit much harder, 30-45% throughput declines reading through on some of the older processor based equipment. Should machine through puts be ranked in the charts using existing best mitigations to give consumers the very best most practical ranking system? Yes, because this Spectre/Meltdown v2 stuff is part of your current reality going forward. Microsoft will blend this mitigation stuff into Win 11 and Win 12, and the Tom's Hardware processor speed re-rankings will flow from that. Intel will try to ignore the whole thing until MS forces mitigations down upon them. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 03/15/22 at 16:17:42 https://liliputing.com/2022/03/lilbits-layoffs-at-arm-new-ryzen-desktop-chips-and-shenzhen-shutdown-could-mean-more-supply-chain-slowdowns.html Just read it, got lots of good rumor/news stuff in it ...... The folks at ARM Holdings are recovering from NVIDIA attempting to take over the world by owning them. ARM finds themselves overstocked by over 15% with a bunch of old NVIDIA employees and now they have to go through the pain and cost to lay them all off. Getting back to a "good competitive posture" as a simple ARM phone chip designer is going to be hard to do as the very best of the ARM employees that they do want to keep can find a better job somewhere else pushing the emerging RISC-V designs. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 03/20/22 at 02:03:10 Apple M2 Ultra chipsets are: Performance superior Cooling superior (lower sized lithography spread over a 4x larger surface means measured temps are actually lower than AMD's temps anywhere you take the Ultra's readings) Graphics superior Still as vulnerable to Specre V2 and Meltdown V2 but with far fewer exploits that have been written against the Mac softwares ........ when compared to AMD or to Intel. Structurally, the Apple M1 Max and M2 Ultra look very much alike EXCEPT the M2 Ultra is physically doubled. Yes, it is two identical M1 MAXs laid out top to bottom on the same substrate. Think Threadripper big or Exynos big if you need something to compare M1 Ultra to. http://https://fdn.gsmarena.com/imgroot/news/22/03/apple-m2-extreme-mark-gurman/inline/-1200/gsmarena_001.jpg The chipset is 4 to 6 times larger than a standard A4 socket AMD chipset, it can pull power like 4-6 of those A4 AMD chipsets and it has the throughput of 4-5 AMD chipsets. Apple has built a series of HUGE BEASTS for their largest processors. Intel no longer has the biggest power draw nor the largest heat produced ---- Apple is now the king of Huge and Excessive. Apple has a working design for a workstation and a mainframe style chipset now ....... =================================================== Intel is a reactive company ....... Intel reacts to what its competitors do (generally by copying them). Intel was all set to compete with AMD. Now Intel has Apple and NVIDIA to deal with. Poor Intel. More and bigger challenges. Intel is now suddenly #4 instead of #2 with a much much much bigger gap in performance to try to overcome. ;) |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 03/22/22 at 11:58:36 https://liliputing.com/2022/03/nvidia-grace-cpu-superchip-is-a-144-core-arm-processor-for-data-centers.html Nvidia shows us what their plans for ARM Hegemony would have looked like if Nvidia had been able to pull off their ARM revolution while completely controlling ARM. Now Nvidia has to compete fairly with Apple and all the others on "a flat and equal playing field" ...... http://https://liliputing.com/wp-content/uploads/2022/03/grace-superchip-780x451.jpg NVIDIA has unveiled a new Grace CPU Superchip for servers that packs 144 ARM CPU cores into a single socket using the same trick Apple used for its M1 Ultra processor: smush two chips together using a high-speed interface. For the M1 Ultra, Apple developed what it calls “UltraFusion” architecture. NVIDI calls its interconnect architecture NVLink-C2C and says it supports bandwidths “of 900 gigabytes per second or higher” while delivering far better efficiency than a PCIe Gen 5 interface. As for the Grace CPU Superchip, in addition to 144 ARMv9 CPU cores, the chip supports LPDDR5x memory with 1TB/s bandwidth and Error Correction Code (ECC). The CPU consumes less than 500 watts when the Grace CPU Superchip ships in the first half of 2023. Another 500 watt plus sized monstrosity created while pulling together another "doubled up" mainframe sized ARM super chip ....... ::) Wow, this one certainly would have would pushed Intel back to #4 position in the new world pecking order ....... I like that 1 terabyte per second throughput bus design though, very good throughput compared to any current PC bus designs ....... |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 03/26/22 at 05:18:18 Intel has just released a new power supply spec that is somewhat revealing. Intel has spec'd their newest class of power supplies has to provide a minimum of 1,000 watts of power. 1,500 watts is included in the spec range on the upper end. Half of this will be likely needed for the Intel graphics, the rest is for the processor and cooling package. Intel is signalling that they will want to play in the new "super huge" brackets right along with Apple and NVIDIA. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 03/27/22 at 08:28:30 https://hothardware.com/news/amd-ryzen-7950x-specs-leak-170w-tdp-16-cores Pre-release rumors from AMD reiterate 12-16 cores that make up distributed chiplet based graphics, higher 170 watt current ratings (covers CPU and graphics functions, remember) and that the Ryzen 7 level will come out hard in advance of the other AMD chipset levels. This uneven roll out is being driven by total chiplet supply issues since the exact same chiplets are used up and down the entire AMD line. The chiplets with built in gaming graphics are going to start out in short supply and these chiplets will be allocated to Ryzen 7 7950x until the supply level increases. AMD will hit the most important, most contested gaming segment first, and they will hit it hard. AMD's next line of desktop processors will be named the Ryzen 7000 series, and they'll be a complete break from Socket AM4 and the previous gen Ryzen platform that it hosts. Ryzen 7000 processors will be based on the Zen 4 CPU architecture and use an LGA socket design rather than the micro-PGA used by Socket AM4. If rumors are correct, the mainstream Ryzen 7000 desktop processors will be based on the "Raphael" design. Aside from Zen 4 CPU cores, Raphael is supposed to include a small set of Radeon GPUs in each processor chiplet, meaning that for the first time, we'll have mainstream desktop Ryzen processors that don't require a discrete graphics card to be used. yep, just like Intel's best products. That integrated GPU is likely contributing to what are rumored to be surprisingly high thermal dissipation requirements for Raphael. We had heard in the past that the Ryzen 7000 series chips would have power ratings up to 170 Watts, despite that they will continue to max out at 12-16 cores on the top-end, just like the Ryzen 5000 and Ryzen 3000 series. Regular leaker Greymon55 tweeted yesterday confirming that information and also adding that there will be Raphael-based Ryzen 9 SKUs at 105W and 65W, too. The top-end 16-core Ryzen 9 7000-series model, likely called something like "Ryzen 9 7950X", will supposedly carry a 170W power rating, while the 12-core model will apparently come with a 105W design power. Several replies in the thread remark that it seems like an extremely high power rating considering that extant 16-core Ryzen 9 chips are rated for 105 W. As Greymon55 points out, these upcoming chips have a lot of things going for them that justify the increased power limit, including higher clock rates, the aforementioned integrated GPU, and brand-new AVX-512 support. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 03/27/22 at 08:35:04 https://hothardware.com/news/amd-ryzen-zen-4-cpus-integrate-radeon-rdna-2-gpus Let's talk some more about those AMD chiplets that each carry their own set of graphics cores and spread over 12-16 chiplets can aggregate together to make up a full gaming level graphics solution. http://https://encrypted-tbn1.gstatic.com/images?q=tbn:ANd9GcQEkNFeppMgYVfdH6DZj_1ocf17NLkR20VfqhqX0LAINTYnx1hl Leaked documents arising from the recent Gigabyte security breach is the gift that keeps on giving. Not that we condone such actions in the slightest, but now that there are interesting tidbits about unreleased products out in the wild, we can't bury our heads in the sand, either. And so with that said, the leaked documents seemingly confirm a previous rumor suggesting all of AMD's next-generation Zen 4 CPUs will have integrated graphics. This is outlined in yet another published document from when hackers pilfered Gigabyte's servers, who then tried extort a ransom from the company. Those efforts were apparently unsuccessful, as the hackers proceeded to publish a trove of stolen data online, containing documents outlining various details about unreleased products from multiple companies, including AMD and Intel. In this case, the folks at Chips and Cheese got their mitts on the data dump and posted a handful of diagrams and other information related to Zen 4. One of the slides details the upcoming AM5 socket that will replace AM4, and what features it will be compatible with, like DDR5 memory and so forth. This is also the chart that suggests all Zen 4 processors will have integrated graphics. http://https://images.hothardware.com/contentimages/newsitem/56051/content/small_zen_4_rdna_2.jpg As you can see above, the chart highlights three types of "Family 19h" processors, which are Zen 4 chips. In each case, it points out on-chip graphics support. So in other words, integrated graphics will not be the exclusive domain of dedicated APUs as part of the Ryzen G-series, but available on all product lineups, to some extent. We say "some extent" because the document also notes, "Some OPNs...may not support GFX." We read this to mean that on some processor SKUs, the integrated graphics will still be there, but will be disabled, just as Intel does with its F and KF processor models. Assuming the information is accurate, this will almost definitely entail the pairing of RDNA 2 graphics with Zen 4. This is something that has been previously rumored, with a supposedly leaked roadmap showing AMD's next-gen CPUs pairing Zen 4 with "Navi2" (RDNA 2). There is still plenty of time speculate. Just a few weeks ago, AMD CEO Dr. Lisa Su reiterated that Zen 4, which is being built on a 5-nanometer manufacturing process, is on track to launch in 2022. Same goes for AMD's next-generation RDNA 3 graphics products. Zen 4 will also bring support for DDR5 memory, just as Intel is doing with its upcoming Alder Lake processors. However, the same set of leaked documents seemingly confirm that AMD will stick with PCI Express 4.0 on the consumer side, and only support PCI Express 5.0 on its next-gen EPYC processor. That really shouldn't matter a whole lot in terms of real-world usage (for consumers), but if that's the case, it does give Intel bragging rights. It's also expected that Zen 4 will hold the line with 12-16 cores and 24-32 threads among the mainstream lineup. Combined with a rumored 20 percent lift in IPC performance, Zen 4 could be a huge upgrade over Zen 3. AMD Zen 4 processors built on TSMC 5nm engineering samples of the chiplet based distributed graphics core Ryzen 7 and the Ryzen 9 are known to be in the hands of board makers and systems builders at this time. 3nm Zen 5 will follow along later in a year or two with its planned core count increases to make the graphics aggregation tricks work even better. 12 -16 core equivalents (or threads if you prefer) is what it takes now (Ryzen 7 at a bare minimum) but this will likely change as AMD rolls down to 3nm lithography late next year. Rumors say the core counts on all levels will roll up to at least the counts used by the next bump up as there is massive amounts of room to do this core count increase inside the same AM-5 socket because of the much smaller more dense 3nm lithography. Both Intel and AMD are going to provide strong gaming graphics across the board going forward. This cannot help but to hurt NVIDIA as it means as long as Intel and AMD stay current on their graphics standards and continue to provide some really good gaming graphics then NVIDIA will lose out on supplying the video cards to these machines. Remember, video cards are one of NVIDIA's main market segments and video cards are the cash cow that makes most of NVIDIA's money for them ...... |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 04/06/22 at 18:01:09 https://www.businesskorea.co.kr/news/articleView.html?idxno=90334 https://www.google.com/search?q=Samsung%E2%80%99s+next+flagships+could+have+MediaTek+inside&oq=Samsung%E2%80%99s+next+flagships+could+have+MediaTek+inside&aqs=chrome..69i57j0i546l2.2449j0j7&sourceid=chrome&ie=UTF-8 After Qualcomm had ditched 4nm Samsung processors for a miserable 25-35% total yield rate we knew there was big trouble brewing in Samsung land. Here is some really major confirmation of that trouble, when Samsung itself begins to use Mediatek brand processors in their very own front line products in preference to Qualcomm processors or their own Samsung designed processors. This wrinkle has never happened before. Apple and Mediatek own the cell phone processing world right now. Simple ARM designs are not the best any longer, the advanced ARM X2 custom designs are the current basis of the winning industry dominant Apple Mediatek in house re-designs. Home made and very much custom designed processors from Apple and Mediatek are leading the phone industry pack right now. Both of these custom design chipsets have customer designed ARM X-2 cores in them for their main horsepower source. http://https://www.businesskorea.co.kr/news/photo/202204/90334_104567_1716.png |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 04/08/22 at 11:51:15 https://www.tomshardware.com/news/amd-future-cpus-could-feature-direct-attached-accelerators http://https://cdn.mos.cms.futurecdn.net/PjhzVvQaaok9JdiGHY3smg-970-80.png.we AMD has patented a processor featuring a machine learning (ML) accelerator that is stacked on top of its I/O die (IOD). The patent indicates that AMD may be planning to build special-purpose or datacenter system-on-chips (SoCs) with integrated FPGA or GPU-based machine learning accelerators. Just like AMD can now add cache to its CPUs, it might add an FPGA or GPU layer on top of its processor I/O die. But, more importantly, the technology allows the company to add other types of accelerators on top of future CPU SoCs. As with any patented work, the patent doesn't guarantee that we'll see designs with the tech come to market. However, it gives us a view into what direction the company is moving with its R&D, and there is a chance we could see products based on this tech, or a close derivative, come to market. Stacking AI/ML Accelerators on Top of Memory on top of a standard socket I/O die AMD's patent titled 'Direct-connected machine learning accelerator' rather openly describes how AMD might add an ML-accelerator to its CPUs with an IOD using its stacking technologies. Apparently, AMD's technology allows it to add a field-programmable processing array (FPGA) or a compute GPU for machine learning workloads on top of an I/O die with a special accelerator port. AMD describes several means of adding an accelerator: one involves an accelerator with its own local memory, another implies that such an accelerator uses memory that's connected to an IOD, while in the third scenario, an accelerator could possibly use system memory, and in this case, it does not even have to be stacked on top an IOD. The rumors progressed from rumors about how AMD could do it, to patents saying AMD has exclusive rights to the AMD based technology to go do it, and now the post patent release rumors extend to proposed release times and the exact lines of AMD product to be affected by the new technology now that the patents have been granted. Intel runs off secrecy and Intel BS while AMD runs off logical chains of conjoined technologies and solid partnerships with other companies like TSMC and XILINX ...... ;) ONCE AGAIN, THE AMD-XILINX-TSMC PARTNERSHIP WITH CONJOINED DISTRIBUTED CHIPLET BASED PROCESSING THAT IS BEING BUILT AND PACKAGED AT THE NEW TSMC FACILITIES USING PROCESSING STACKS THAT ARE TO INCLUDE THINGS LIKE TSMC IN-CHIP LIQUID COOLING. THIS OFFERS SOME REALLY REALLY MASSIVE ADVANTAGES TO AMD, GIVING AMD SOME WORLD CLASS PROCESSOR COOLING TECH THAT INTEL ALSO REALLY DESPERATELY NEEDS (but doesn't have) FOR THEIR OWN SUPER HIGH WATT DRAW INTEL ROOM HEATERS. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 04/09/22 at 18:47:07 Now we have looked at something huge and complex, let's go check out the simple and cheap end of the pond ...... http://https://liliputing.com/wp-content/uploads/2022/01/mq-pro_06-780x491.jpg The MangoPi MQ Pro is a single-board computer that looks like a Raspberry Pi Zero, but instead of an ARM-based processor, the MQ Pro is powered by a RISC-V chip. MangoPi first announced it was developing the MQ Pro in January, and now it’s available for purchase in China for about $20. It should be available worldwide soon. At the heart of the little computer is an Allwinner D1 processor, which is the same 1 GHz, single-core, 64-bit RISC-V processor we’ve seen used in other small cheap devices including the Sipeed Lichee and Nezha. Like a Raspberry Pi Zero, the little computer measures 65 x 30mm (2.6[ch8243] x 1.2[ch8243]) and features a few ports including two USB Type-C ports, a mini HDMI 1.4 port, and a microSD card reader plus a 40-pin Raspberry Pi-compatible GPIO header. While that’s a pretty limited set of ports, the Mangopi MQ Pro could also work with carrier boards that will add features like full-sized Ethernet and USB Type-A ports. MangoPi will offers versions of the MQ Pro with 512MB or 1GB of DDR3L memory, and the little computer supports WiFi 4 and Bluetooth 4.2 wireless connections. There’s some documentation for the MQ Pro at the MangoPi website and the company’s GitHub page, including schematics, datasheets, tools, and suggested firmware (the board will be compatible with Tina-Linux, but as of early April, 2022 the firmware isn’t available for download yet). The MQ-Pro will be MangoPi’s second mini PC with a RISC-V processor. The first was the even smaller 1.6[ch8243] x 1.6[ch8243] MangoPi MQ, which has an Allwinner D1s processor, 64MB of RAM, two USB-C ports, a microSD card reader, WiFi and Bluetooth, but no traditional video output port (although there is a 15-pin DSI FPC connector). Currently limited to just one 64 bit core and at a price that fluctuates between $10-20, one wonders what the upcoming 4 core versions of this will be able to handle. Tina Linux (developed by Allwinner specifically for the D1 board products) isn't a widespread powerhouse of an operating system, even when compared to Raspian (built for the Raspberry Pi products). Right now any older Raspberry Pi is a better unit with much more fully developed software. Raspberry Pi 4 can better fake being your main PC "more acceptably" if you need it to do so. So can any old Chromebook that exists anywhere in your family. ;D At $163 Raspberry Pi is no longer quite as compelling as it used to be. https://www.amazon.com/seeed-studio-Raspberry-Personal-Including/dp/B08W2M2MMM/ref=sr_1_3?crid=319RTV60SDB15&keywords=raspberry%2Bpi%2B400&qid=1649648967&sprefix=raspber%2Caps%2C85&sr=8-3&th=1 Don't look to Broadcom to supply a new chip to Raspberry Pie this year. Broadcom just got bought out by Avago two years ago and is right now being combined with two other more recent tech Avago purchases. So, Broadcom is still being REFOCUSED on what Avago wants it to build. Alton Upton and Raspberry Pi guys are seen as a "past casual hobby" by the new Avago owners, who will soon stop building the Pi processors just as soon as industry demand for the base chipset drops off. Broadcom in general is losing their bottom end products to RISC-V just as soon as a good enough RISC-V design gets built. Expect Alton Upton and crew to switch over their processors to RISC-V as soon as this is feasible to do as Upton's old big discounts at Broadcom are becoming a past tense thing as well. |
|
Title: Re: AMD & others --- Intel dominance in 2022 ? Post by Oldfeller--FSO on 04/19/22 at 11:49:43 https://www.tomshardware.com/news/amd-ryzen-pro-6000-series-specs http://https://cdn.mos.cms.futurecdn.net/wdvCwW4PUaof4Ps3tyfJd7-970-80.jpg.we Ryzen 6000 at 6nm will sell for at least two quarters, then will be replaced by 5nm Ryzen 7000 units which will phase into the line to replace the 6000 series units. 5nm Ryzen 7000 will have "up to twice as many chiplet cores" and these new 5nm chiplet cores will have the new built in RD-3 gaming graphics which will render it a clearly superior product in a laptop. Intel can compete with Ryzen 6000, but only at power draw levels that are clearly much higher and much less desirable in a laptop product. First word is Intel will be running 2x higher current draw for Intel to try to get same level performance as a Ryzen 6000. Intel currently has no disclosed Intel 10nm built competitive unit for Ryzen 7000, but Intel says Intel will begin to beat and compete again when they start building 3nm chipsets at TSMC two years from now. Two years is forever in computer terms. In two years, ASML will be building new generations of 2nm and 1nm capable scanners. TSMC and Intel will be installing them in American native soil complexes as China will have likely started to Biden'd their way into Taiwan and, if rumors are true, the Taiwanese ASML scanners will have been moved to America or else will have been blown up in place ahead of a Chinese take over. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 04/25/22 at 08:01:34 Software based Mitigations for Spectre Version 2 are out now native in Ubuntu and Linux Mint formats. I have the slowest Core 2 Duo machine that is available used from Dell, so if anyone is suffering it should be me. Boot times include boot scanning now, and it shows. It takes 5-7 seconds longer to boot from a cold boot. Opening programs takes 1-3 seconds longer. Programs get scanned when first started up. Communicating over web based anything takes an extra second or so to get it started up. Point made earlier about Intel effectively losing their last big performance bump to mitigations slowdowns seems to have panned out when using my old Intel box as the baseline. ================================================== There are tricks to do to avoid most of this slow down at start-up. Just turn your monitor off and leave your CPU running all the time in idle mode sending info to a monitor that is simply shut off. MS has been doing this trick for a decade now, as their junk was way way too slow way way back then ...... =================================================== New Intel based Dell Alder Lake laptops are out. Instead of 15-16 watts, they pull 28 watts. Battery life sucks accordingly, but that is what you have available to buy right now as the new generation AMD units have all been snapped up and are GONE already. The AMD decks have been cleared for the Ryzen 7000 series release starting next month ....... AMD's next gen product will be the 5nm stuff with the extra cores and extra stacked on top memory and the self-aggregating built in gaming graphics ....... This will come out in the gaming range of Ryzen 7 pulling like 12 cores and simply using up AMD's entire wafer allocation right off the bat. AMD's total rewrite of everything that will be called Ryzen 7 comes out in spots starting next month. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 05/01/22 at 20:22:39 http://https://cdn.mos.cms.futurecdn.net/QsQyNEqDQx4ENvo7FFh3y9-970-80.png.we OK, AMD's 6nm refresh with the big extra memory lands on the top of the chart leaving us waiting for Intel's counter moves. AMD is not going to fill in all their units in the line as they will be replacing it all with Ryzen 7 very shortly. ;D The old big swashbucking swordfight up the big set or stairs analogy still applies to what is going on this summer and fall. Fighting continuously while moving up together a step or two at a time ........ All the new stuff from everybody needs to rank 200 or better to even exist going forward from here. Intel still pulls double the wattage of AMD on all new Intel items ---- This makes up a very real and long term killing disadvantage to Intel going forward. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 05/04/22 at 06:57:20 https://hothardware.com/search?a=all&s=amd+reveals+dragon+range The AMD earnings release also incorporated news about new Ryzen 7000 mobile chipsets. http://https://images.hothardware.com/contentimages/newsitem/58449/content/small_ryzen-7000-dragon-range-slide.jpg AMD Zen 4 Dragon Range, Phoenix, And Xilinx Integration In addition to the financial disclosures, AMD also revealed some information regarding future products. In her prepared remarks, AMD CEO Dr. Lisa Su mentions future processors with integrated Xilinx technology. “As one example, we are integrating Xilinx’s differentiated AI engine across our CPU product portfolio to enable industry-leading inference capabilities,” Dr. Su said, “with the first products expected in 2023.” Ryzen 7000 Dragon Range slide Along with that news, AMD’s Robert Hallock also officially announced two new Zen 4-based Ryzen 7000 series mobile processors – Dragon Range and Phoenix. We caught wind of Phoenix quite a while back, but Dragon Range was a bit of a surprise. Both Phoenix and Dragon Range will leverage AMD’s upcoming Zen 4 architecture. Both chips will offer PCI Express 5 and support for LPDDR5 memory, though it wasn’t clear if DDR4 would also be supported. Phoenix will have a 35-45 watt TDP range and target thin-and-light gaming notebooks with z-heights less than 20mm. Dragon Range, however, will be a more powerful CPU, with a 55W+ TDP range, for enthusiast-class, high-end gaming and creator-focused notebooks. Hallock also noted that Dragon Range will have the highest core and thread counts, and the most cache, for a mobile gaming processor, but actual specifications and details weren’t disclosed. =================================================== Key details from AMD’s financial result disclosures were also covered: Q1 Results Record quarterly revenue of $5.9B, an increase of 71% y/y and 22% q/q Street revenue estimate for Q1 based on analysts who have updated their models to include the partial quarter of Xilinx is $5.5B Gross Margin of 53% An increase of 660 bps y/y and 240 bps q/q driven by higher EPYC server processor revenue and addition of Xilinx Record quarterly non-GAAP net income was $1.6B Up 148% y/y and 42% q/q Record quarterly non-GAAP EPS was $1.13 Up 117% y/y and 23% q/q Excluding the Xilinx additions, AMD’s Q1 results are still strong: Revenue increased 55% y/y and 10% q/q to $5.3B We guided revenue to increase 45% y/y to $5B Gross margin was 51%, up 480 bps y/y and XYZ bps q/q We guided to 50.5% AMD also provided some updated guidance that incorporates new business as a result of the Xilinx acquisition. To quickly summarize, AMD expects Q2 revenue to be approximately $6.5B (+/- $200 million), which is an increase of approximately 69% year over year. AMD expects full year 2022 revenue to be approximately $26.4 billion, an increase of approximately 60% over 2021, due in no small part to the increased margins associated with Xilinx’s higher server and semi-custom revenue. This is a massive increase from prior guidance of approximately 31%. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 05/04/22 at 08:40:42 Confirmation of earlier AMD information showing that AMD will compete against Intel's proposed higher current draw main processor "assbuster" specifications and also against Apple's more moderate draw mobile chipsets using the same chiplets set up in different chiplet counts and memory configurations. https://liliputing.com/2022/05/amd-ryzen-7000-dragon-range-chips-for-gaming-laptops-coming-in-2023.html AMD plans to launch the first Ryzen 7000 series desktop chips later this year, with mobile versions aimed at laptops and other compact computers coming in 2023. But that oversimplifies things a bit, because it turns out AMD will actually be launching two different lines of laptop chips next year. In addition to the previously expected “Phoenix” chips aimed at thin and light gaming laptops, AMD will launch a line of “Dragon Range” chips for higher-performance gaming laptops. http://https://liliputing.com/wp-content/uploads/2022/05/ryzen-7000-780x418.jpg In a nutshell, both chips will use the same Zen 4 CPU cores as the desktop-class “Raphael” processors set to launch this year. They’ll both also support PCIe 5. But Dragon Range chips consume more power and are expected to have higher core and thread counts and/or speeds than Phoenix chips, more cache, and overall higher performance. They’re also expected to show up in laptops that are more than 20mm (0.78 inches) thick, suggesting it’ll take more cooling power to keep them happy. Basically, these are chips for the kind of laptops where performance is more important than battery life, and where you’ll probably keep them plugged in running off charger power most of the time. In a nutshell, here’s what the Ryzen 7000 lineup looks like so far: Raphael: 65W+ desktop chips with Zen 4 CPU cores and PCIe 5 support Dragon Range: 55+ laptop chips with Zen 4 CPU cores and PCIe 5 support Phoenix: 35-45W laptop chips with Zen 4 CPU cores and PCIe 5 support Note that the in the graphic shared by AMD, some chips are described as supporting DDR5 while others show LPDDR5, but Ian Cutress notes that this doesn’t mean they only support those types of memory. So don’t be shocked if a laptop with a Phoenix CPU ends up featuring DDR5 memory or a Dragon Range system supports LPDDR5, for example. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by LANCER on 05/04/22 at 10:49:01 Well, ok then. |
|
Title: AMD & others --- Intel dominance in 2022 ??? Post by Oldfeller--FSO on 05/06/22 at 04:56:55 Got the first wave of AMD new generation laptop all in one processor reviews coming in now. AMD's very least 6nm unit is a Chromebook processor. Least amp draw, least throughput. This new 6nm Chromebook processor out powers the existing (previous generation) high watt draw laptop processors and even some of the least of the desktop processors that are currently selling. AMD 5nm Ryzen 7000 may rule for a while as Intel has no answer for it at all until TSMC makes them some 3nm chipsets on TSMC's equipment. ================================================== Please remember, AMD must concentrate their newest 5nm and 6nm chiplets where they will do the most good. AMD is still going to be 5nm wafer limited and AMD will remain wafer limited as Intel rolls further into TSMC's supply allocation system. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 05/10/22 at 12:37:35 https://liliputing.com/2022/05/intels-55-watt-alder-lake-hx-chips-bring-up-to-16-cores-24-threads-to-laptops.html http://https://liliputing.com/wp-content/uploads/2022/05/hx_02-780x436.jpg Intel strikes back at AMD's top of the chart laptop offerings, quadrupling Intel's base rated current draw numbers, lasering off some built in graphics support and killing all of Intel's much ballyhoo'd Thunderbolt 4 support with all this done in an attempt to keep the current draw numbers from being so frighteningly durn AMAZINGLY HUGE. Another read is that Intel has some serious production issues that mean advanced graphics segments and Thunderbolt subsections simply aren't working right at all and get laser truncated (so Intel tries to make a feature out of of the lack, huh?) These newest Intel laptops can still draw more power than Intel desktops used to draw ...... and they are NOT FEATURE COMPLETE nor are they even literal competitive products compared to last year's Intel even. Like all of Intel’s 12th-gen chips, the new processors are based on a hybrid architecture that combines Performance (P) and Efficient (E) CPU cores. What’s new for the HX series is: Processor Base Power of 55 watts (up from 45 watts for Alder Lake-H chips) Max Turbo Power of 157 watts (up from 115 watts for Alder Lake-H) Support for up to 8 P cores and 8 E cores (up from 6P + 8E for Alder Lake-H) Intel Xe-LP integrated graphics with 32 execution units (down from 96eu for Alder Lake-H) Thunderbolt 4 support is no longer integrated with the chip support for PCIe Gen 5. In other words, you get more cores and more power directed to CPU performance, but it comes at the expense of integrated graphics. That makes sense though, as you’re unlikely to find a laptop with an Alder Lake-HX chip that doesn’t have discrete graphics, since these chips are designed for high-performance gaming laptops and mobile workstations. The loss of integrated support for Thunderbolt 4 also means that Thunderbolt will be an optional feature for laptop makers, since they’ll have to add a discrete controller for that functionality. Again, given the target market for these chips, I suspect that we will see Alder Lake-HX laptops with Thunderbolt support, but you can’t take it as a given that if a notebook has one of these new chips that it will also have a Thunderbolt port. http://https://liliputing.com/wp-content/uploads/2022/05/hx_03-780x442.jpg Intel says the new Core HX chips will be the final entries in the 12th-gen Intel Core family before the company moves on to 13th-gen processors. For now, there are 7 different chips in the Alder Lake-HX lineup: Cores Threads BaseFreq (P+E) Turbo Freq (P+E) L3 Cache GPU TDP Current draw (basic and turbo modes) Core i9-12950HX 16 (8P + 8E) 24 2.3 / 1.7 GHz 5.0 / 3.6 GHz 30 MB 32 EU / up to 1.55 GHz 55 W 157 W Core i9-12900HX 16 (8P + 8E) 24 2.3 / 1.7 GHz 5.0 / 3.6 GHz 30 MB 32 EU / up to 1.55 GHz 55 W 157 W Core i7-12850HX 16 (8P + 8E) 24 2.1 / 1.5 GHz 4.8 / 3.4 GHz 25 MB 32 EU / up to 1.45 GHz 55 W 157 W Core i7-12800HX 16 (8P + 8E) 24 2.0 / 1.5 GHz 4.8 / 3.4 GHz 25 MB 32 EU / up to 1.45 GHz 55 W 157 W Core i7-12650HX 14 (6P + 8E) 20 2.0 / 1.5 GHz 4.7 / 3.3 GHz 24 MB 32 EU / up to 1.45 GHz 55 W 157 W Core i5-12600HX 12 (4P + 8E) 16 2.5 / 1.8 GHz 4.6 / 3.3 GHz 18 MB 32 EU / up to 1.35 GHz 55 W 157 W Core i5-12450HX 8 (4P +4E) 12 2.4 / 1.8 GHz 4.4 / 3.1 GHz 12 MB 16 EU / up to 1.3 GHz 55 W 157 W If you’re wondering why some of those chips have virtually identical specs, it’s because models that end in 50HX are aimed at enterprise customers and include Intel vPro features. This is the last gasp for lithography equipment built and designed in house by Intel. From this point going forward, all production equipment will be designed by ASML (Dutch) and designed for production at TSMC (Taiwan). Intel as you knew it is over now ....... 10nm was as small as Intel ever got to run at all and it is clear that scrap rates are overcoming that native Intel 10nm product line at this point. There is really only one i5, one i7 and one i9 final 12th gen Intel design CURRENTLY in reality, just one set designed for laptop and one other set for desktop. All the variants you see listed in this table come from selectively lasering off non-functional graphic sets, non-working i/o systems and misc. other items that are simply not working right just in an attempt to curtail the out of control current draw these last Intel designed chipsets represent. In the two years Intel has to wait for TSMC to pick them up with a good product on a smaller, more modern lithography you will undoubtedly see several "new, improved, latest, greatest" variations from Intel that are actually lasering stuff off so as to sell the sorted scrap as a processor. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 05/10/22 at 14:36:06 https://liliputing.com/2022/05/lilbits-mips-pivots-to-risc-v-intel-stagers-desktop-arc-graphics-release-and-google-releases-i-o-pinball.html Also today, MIPS released their plan to completely drop their own MIPS standard and to adopt and to technically support RISC-V for all processor sets going forward. MIPS might be able to make up a good RISC-V core using their extensive knowledge and that would do more for reversing their corporate fortunes that any amount of old MIPS designs that are going absolutely nowhere right now. http://https://liliputing.com/wp-content/uploads/2022/05/evocore-723x500.jpg Last year MIPS Technologies announced that it was going to stop designing MIPS processors. That might sound surprising given the company’s name, but maybe it was inevitable when looking at trends in the semiconductor industry. So MIPS pivoted to RISC-V architecture. And the company’s first chips based on that open instruction set are set to launch later this year. MIPS is promising best-in-class performance, but we’ll likely have to wait until this fall to find out whether the company can deliver on that promise. What this is good for is to see a RISC-V typical PC class design system run out into the particulars. It is dirt simple on the chiplet level, memory is local and is small per chiplet. Chiplets aggregate into an as yet unnamed "cluster something" that can run a function, then the cell design multiplies that "cluster something" a bijillion times to get larger processing capability. Reminder, MIPS wasn't a world leading company there at their end, and their take on how RISC-V is going to be structured in the long run isn't absolute gospel for anybody but them at this point in time --- so take this structure with a goodly dose of salt. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 05/13/22 at 19:08:17 OK, Intel is not dominant any longer in processor design and manufacture. As of last month, Intel is not even a leading player using anything they have designed or built for themselves. Intel is also getting increasing amounts of pushback on their various claims of having made "massive deals" with TSMC and ASML. All Intel has done is shift all the real processor producers in Taiwan and South Korea over to a philosophy of "bad customers pay upfront in total, including any anticipated scrap rates" production allocation basis for all normal customers. Apple has a lower cost at TSMC, based on being on board from new lithography concept to production to moving on over to the next new thing. Apple generally pre-pays for the first lines of a new lithography level and uses those lines exclusively until they move on to the next new one. AMD has a moderately discounted cost because they order well in advance and pay all the requested amounts up front to place an order. AMD moves in when Apple leaves a process level, getting a cost based on a paid for known stable production system that is well known and debugged. Intel pays retail, up front, with larger scrap rate allowances as they attempt to buy into the leading edge of production lithography. If Intel cannot overcome their very rocky relationship issues with TSMC, Intel will not survive the next decade. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 05/13/22 at 19:29:03 https://www.tomshardware.com/news/amd-share-skyrockets-amid-biggest-quarterly-desktop-pc-decline-in-history http://https://cdn.mos.cms.futurecdn.net/zhKTqkB8xfD9MizrxMxsCX-970-80.png.we Surprisingly, AMD managed to carve out significant wins during the tumultuous quarter and AMD has now, once again, set a new record high x86 market share of 27.7%, an incredible increase of seven percentage points over last year. Both Intel and AMD suffered from the disturbingly fast decline in the desktop PC market, but AMD didn't lose sales quite as quickly as Intel, resulting in a share gain for the quarter. Notably, much of the decline in desktop PC came as vendors burned through excess CPU inventory at high discounts, which McCarron says impacted Intel more severely than AMD. As such, Intel still gained some unit share in the desktop PC market compared to a year ago, but the gain was quickly overcome by the general decline in the desktop market. AMD CEO Lisa Su recently predicted that the company would also see a decline in the 'high single-digit' percentage for the desktop PC TAM for the year, indicating that shrinking desktop PC sales could continue. AMD would move chiplets from PC over to laptop and enterprise accordingly and AMD would not have to run a fire sale on completed processor sets like Intel is currently doing. AMD continued to take big strides in the mobile/laptop market as it set another record for unit share in that segment with 22.5%. AMD also gained in the server market for the 12th consecutive quarter, reaching 11.6% of the market. So, when Intel bogus claims big gains in PC, they are really claiming they lost ground less fast than what was originally thought. This is not real growth at all. At best, Intel gained at most one temporary percent point and that was short lived at best. AMD however has ground out an ongoing very real 7-21% growth in some very bad times making up the growth shown in these non-PC segments, growth results which reflects AMD's strong market position and AMD's very sound products. =================================================== Intel Meteor Lake details are now surfacing ....... Intel will use two classes of Intel designed and built chiplets based on 10nm and 7nm but will be using their new deceptive nomenclatures and will be using the new Intel power draw metrics that HIDE the real current draws (which are very high) under the newly re-defined Intel current draw metrics. Intel is absolutely counting on you NOT ACTUALLY KNOWING WHAT YOU ARE BUYING. Not that Joe and Rita Sixpack are really caring very much about it either. ::) Intel is currently having a hard time moving their existing inventory of finished PC class products. AMD is selling through at full price much better compared to Intel as AMD's honest metrics that people can understand are helping AMD to sell their wares. Intel is already marking down their pre-built PC boxes greatly for the new school year ....... BIG SUPPOSE by Intel is that they can get a usable yield off their old 7nm process, and failing that Intel will place orders on TSMC (and wait) for what they need. Intel is seeing a future controlled by TSMC allocations just like AMD sees. TSMC sees building some more American facilities as the US gov requires their chipsets to be made on US soil. This is slow and expensive to accomplish, and guess who gets to pay for it ........ Samsung is attempting to overcome their recent stumbles on bad processor yields and get a customer to order some processors from them. So far, no luck. |
|
Title: Re: AMD & others --- Intel dominance in 2022 Post by Oldfeller--FSO on 05/17/22 at 23:14:18 New rumors about AMD's 3nm chipsets are coming out in the info flow from oriental suppliers. Ryzen 3000 Matisse 7nm (Zen 2) 16/32 105W AM4 500-Series DDR4-3200 Gen 4.0 2019 Ryzen 5000 Vermeer 7nm (Zen 3) 16/32 105W AM4 500-Series DDR4-3200 Gen 4.0 2020 Ryzen 5000 3D Warhol? 7nm (Zen 3D) 8/16 105W AM4 500-Series DDR4-3200 Gen 4.0 2022 Ryzen 7000 Raphael 5nm (Zen 4) 16/32? 105-170W AM5 600-Series DDR5-5200/5600? Gen 5.0 2022 Ryzen 7000 3D Raphael 5nm (Zen 4) 16/32? 105-170W AM5 600-Series DDR5-5200/5600? Gen 5.0 2023 Ryzen 8000 Granite Ridge 3nm (Zen 5)? TBA TBA AM5 700-Series? DDR5-5600+ Gen 5.0 Name for the AMD 3nm stuff is Ryzen 8000 Granite Ridge Intel keeps on frothing about 24 cores (some big some little). AMD does not see a lot of advantage in lots of little cores once the main core count goes over 10 as minor tasks can be allocated to an idle larger processor just as easily. When AMD posts core counts of 16 they are talking larger main performance cores that can support 2-4 threads per core. Intel little cores can only do one thread at a time. Next questionable item is how many threads per core is going to be available as AMD has tested up to 4 threads per core ..... but once again AMD did not see much real advantage in the higher thread per core counts right now under existing Microsoft OS systems. This is also why Linux systems test better than MS systems do right now. This is another reason MS sux compared to Linux --- especially in more powerful machines. If Microsoft creates an OS that actually uses processor threads better, this could all change up fairly rapidly. AMD has upped the speed of their newest cores to 5.4 gigahertz --- this along with a memory system and an I/O system that is equally fast means better computing throughput overall will be available. When Intel is behind in a product class (and it so definitely is right now) Intel just copies the leader of that class in great detail. Intel has started copying Apple mobile product core designs and Intel is copying AMD's chiplet design patterns for consumer PCs as well. Intel has lots of legal resources, and once shown a better way forward quickly tries to make that path their own. Intel will immediately try to patent your better idea or something very close to it then sue you (the real originator) for patent infringement and then settle out of court for "shared rights". AMD's current Threadripper product line is getting lapped by AMD's normal line up and by Intel's 24 core products so expect it to go away soon. It may be back later with a new generation of processor chiplets (and when AMD has enough spare chiplet allocations to give to it). =================================================== https://www.digitaltrends.com/computing/amd-zen-architecture-roadmap-leaked/ Full AMD Zen road map leaked, giving a peek of what’s coming By Monica J. White May 13, 2022 We’re still months away from the launch of AMD Ryzen 7000 “Zen 4” processors, but new leaks tell us more not just about Zen 4, but also about its successors: AMD Zen 5 and Zen 6. Although there’s plenty to be excited about with the Zen 4, AMD seems to have even bigger performance jumps in store for us in the future, if the rumored architecture of Zen 5 is anything to go by. This massive round of leaks comes from YouTube channel Moore’s Law is Dead. Citing anonymous sources, the YouTuber talked about what we can expect in terms of architecture from the upcoming AMD Ryzen chips. This includes both consumer-level chips (AMD Zen 4, Zen 5, and Zen 6) as well as Zen 4C, which is aimed at data centers. The upcoming AMD Ryzen 7000 Zen 4 CPUs are said to retain a similar architecture to Zen 4 and will bring an upgrade in terms of cores and clock speeds. Zen 4 may bring a significant instruction-per-clock (IPC) boost, ranging from 15% to 24%. Single-threaded operations may see an uplift of 28% to 37%, and the same will be true for multi-threaded performance, although the boost may be even bigger. As mentioned, clock speeds will definitely be higher in the Zen 4 lineup than they are in AMD’s current arsenal. Twitter user Petykemano has recently spotted what appears to be a very early benchmark of the AMD Ryzen 7 7800X. The processor hit 5.2GHz, marking a significant upgrade over its predecessor — the Ryzen 7 5800X can only hit up to 4.7GHz. However, it appears that the Zen 4 CPU retains the same number of cores as the Ryzen 7 5800X, appearing with eight cores and 16 threads in the benchmark. Keep in mind that the listing has since disappeared, so all of this is subject to change. Zen 4 is also said to double the size of the L2 cache while keeping the L3 cache the same as in Zen 3. Another huge change for AMD, which is not news at this point, is the fact that the processors will finally support PCIe 5.0, DDR5, and LPDDR5. This will remove the edge Intel Alder Lake has had over AMD throughout the year, seeing as Zen 3 doesn’t support these technologies. On the downside, the rumor mill has been buzzing with the news that Zen 4 may have no DDR4 support whatsoever. AMD Ryzen 7000 Zen 4 rumors. Moore's Law is Dead Lastly, the Zen 4 lineup will cover a vast range of processors. The DIY market will receive a full range of Ryzen 7000 Raphael CPUs sometime in the second half of 2022. Laptop users have AMD Ryzen 7000 Dragon Range and Phoenix to look forward to, with both set to release in early 2023. There are also AMD EPYC Genoa 7004 processors set to launch in the fourth quarter of 2022 and Threadripper 7000 Storm Peak in the first half of 2023. Following Zen 4, AMD will release the 5nm-based Zen 4C, which will target data centers. While the AMD Genoa will utilize Zen 4 cores, the follow-up AMD Bergamo will switch to Zen 4C, bringing the core count up to 128 cores (whereas Genoa will max out at 96). Bergamo will also offer up to 256 threads, and it will be compatible with the same LGA 6096 socket as its predecessor. We can also expect 12-channel memory support. We’re done with Zen 4, but there’s more to talk about, namely the future. AMD Zen 5 is still a while away, but all the sources cited by Moore’s Law is Dead expect it to arrive only 11 to 15 months after the launch of Zen 4. While that’s a short time, the CPUs are said to bring major upgrades, with a performance leap as big as Zen 2. Zen 5 is said to come with a full architectural redesign, resulting in a huge IPC boost, seemingly larger than what the switch from Zen 3 to Zen 4 will bring. On the other hand, the clock speeds will not change much, but we will see changes made to the data fabric and a fully reorganized cache design. AMD’s Zen 5 processors are rumored to be based on TSMC’s N3 or N4P process node. AMD Zen 5 and Zen 6 leaks. Moore's Law is Dead The first Zen 5 product line to ship may be the EPYC Turin CPU, set to arrive in the second half of 2023. The consumer-level products won’t appear for a while, with Ryzen 8000 Granite Ridge (desktop) expected in late 2024 and Ryzen 8000 Strix Point (laptop APU) in early 2025. As for AMD Zen 6, we’re still a long while away from its launch, and by the time it’s released, it may not even be called “Zen” anymore. The naming convention seems to be a bit up in the air, but things on the specification front are quite exciting, including higher core counts and clock speeds, a new cache design, new accelerators, and more. We got Intel going silent as AMD laps their best big little 10nm stuff. Intel desperately needs a fresh new lithography and is asking TSMC to provide it for them at TSMC 3nm ---- which is years away at this point in time. See AMD continue taking even more market share from Intel with this action to continue until Intel has a for real better product to sell. Intel's response is to put out "more Intel marketing" which smells like more puffs of brown vapor and hidden higher current draw BS run off of 10nm and eventually on Intel 7nm whenever/if ever they get it running right. AMD is now running chiplets off 5nm nodes at TSMC right now for real and is shipping sample processors built with them. |
|
SuzukiSavage.com » Powered by YaBB 2.2! YaBB © 2000-2007. All Rights Reserved. |