Donate!
Welcome, Guest. Please Login or Register :: View Members
Pages: 1 2 3 4 ... 31
Send Topic Print
2020 -- new Intel failures & successes (Read 12299 times)
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & issues
Reply #15 - 06/30/18 at 08:03:23
 

https://www.barrons.com/articles/intels-optane-chips-getting-closer-to-changi...

More BS Intel PR being reported yet again, Optane isn't going anywhere except in Intel's wet dreams .....




OK, this is compare and contrast time.   Intel Optane vs Samsung, Micron and all the rest.

In 2015, Intel made a big noise about Optane (saying the whole world would be running on Optane "phase change" memory inside of six months) and Intel has actually sold some Optane stuff in the last 2 years, but it has only successfully been used as a cache memory of sorts in a few very large rack space applications because Intel found they had early "read/write dependability" slowdown issues and a rather severe cost constraint acting against them.

Then Micron jumped ship on the original Optane idea and left the partnership completely this past spring.

Micron simply moved over to the Samsung side of things and threw their support behind the newer SD Association standards (currently at level 7) for both the SD/UC and SD/Express standards.    Micron has mentioned some vague plans to do something with Optane II down the road (Optane II which is still completely undefined at this point in time).  

Intel will have no rights to produce Optane III (as it will be designed by Micron at that future point in time).   Optane III will require a new brand name then, too, as Intel does share rights to the word "Optane" and Intel has so abused and tarnished the image of the "Optane" brand name that it is unrecoverable at this point.   Optane version 2 is muddled, and Micron won't finish working on it until Intel pays for all the Optane they had contracted for last year.

So, this can mean a "complete fuzzy something" may come leaping up at the world down the road because "Optane" as a term and as a standard is really truly completely undefined and totally unexplained at this point in time.

Literally anything could be "Optane" to Intel's eyes .......  so please, buyer beware on Intel Optane stuff going out into the future.   You really DON'T know what you are getting.

Impartial 3rd party tests of existing Optane set ups so far show that it does not yield the real performance levels that Intel advertised  to sell it to you in the first place.


===================================================


So, Intel is the only one producing any of the original Optane at 14nm only at this point in time.    Micron, Samsung and the rest are running the new SD Association SD/UC standard at 12nm and at 10nm .....  and have plans to run the SD/Express level at 7nm next year (Samsung is currently starting early memory production at 7nm as we speak so you can easily see where this is headed inside the next 2-3 years).

Same general use limitations apply to both Optane and to SD/UC and SD/Express at this point in time --- it really isn't fast enough to run as main systems memory just yet.   But it is faster than the older SSD drive tech and certainly faster than any sort of platter hard drives.

Cost is still a real issue when you start talking anything about "Terabyte sized" anything in micro SD card sizes.   Micro SD cards really are quite expensive once you go past 128 gigabytes heading up towards 258 gigabyte and 516 gigabyte sizes which are all that is available right now to be bought in SD/UC levels 1,2 & 3 on a Micro SD format.

These SD/UC levels 1, 2 & 3 cards are somewhat confusing right now, but they are REAL PRODUCTS and because of the current wide acceptance, lower cost and good availability these standards will be the eventual winners over the original Optane from Intel.

And yes, you could put your whole hard drive on a multi terabyte micro SD sized card swinging the SD Express double row of contacts "standard socket" starting early of next year.

You would need a device swinging the new SD/Express double row socket and some fairly deep pockets, but yes, you could go do it.

https://www.cnet.com/news/new-sd-card-uc-express-specs-tout-128tb-max-up-to-a...

Point to consider, is that when 7nm production actually begins to shift down to 5nm (in 2-3 years from now this may actually happen) we will likely be at the SD Association's level 8 standards, and will still only be using less than half of the total capabilities that the new level 8 standard provides.

However, right now the Micro SD/UC grade cards at 10nm (grades 1-3) are readily available and for $18 shipped I found I could buy a 64 gigabyte card that supposedly is just as fast as my new phone can supposedly actually use.  

Android 8.1 is the first OS level that will supposedly really allow the card to "conjoin" with the 32 gigs of systems memory that comes on the Moto G6  in what is again promised as "a user seamless manner" under the control of an updated Memory Manager (newly re-implemented in Android 8.1).       We shall see, my new 64 gig Micro SD/UC-1 card gets here this afternoon.

NO, IT IS NOT REAL YET

        Roll Eyes       Nope, it is the same old SD stuff, music, pictures and maps and only a very few apps can move over to the SD card at this time even if the SD card is custom formatted and encrypted as "internal" and the Memory Manager is active.   Memory Manager simply automatically takes the least used items off your systems memory as it starts to get full and puts those items over on to the SD card.


My past experiences with Android phones being able to actually "seamlessly and completely use" the SD card memory WERE NOT REAL on all my past phones and I had hoped that this lack had finally changed for the better.   Nope, exactly the same restrictions are on the SD card use as before.  Pictures, music, graphics and a few permittable apps are the only things moved over at this time .....  HOWEVER, there is a detailed set of folders created on the drive by the Memory Manager that could be filled up with properly sorted and filed items but that will only happen as your normal memory begins to clog up with stuff.

The new "super card" installed as conjoined internal memory cannot be speed tested by any of the the old standard android SD card speed tests as they report it correctly as custom encrypted memory and they never were built to handle that sort of action.

Tongue    ........ oh well, I can't even test it to see if it is working right.
Back to top
 
« Last Edit: 08/07/18 at 14:42:09 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & issues
Reply #16 - 07/02/18 at 06:41:41
 

https://www.extremetech.com/computing/270550-are-arm-cpu-cores-finally-ready-...

Are ARM CPU Cores Finally Ready to Fight
Intel for the Laptop Market?




Last week, ARM announced its next-generation CPU architecture, the Cortex-A76. We’ve already covered the announcement and initial performance expectations, but in the wake of some additional deep dives and positional information, it’s worth revisiting whether this is finally the CPU that’ll take on Intel in the laptop market.

If you listen to ARM, it absolutely is. Mike Filippo, ARM’s chief architect, told CNET that he expects the Cortex-A76, which should be in-hardware by 2019, to have roughly the same performance as a Core i5-7300. Outfitted with more cache, Filippo expects that the Cortex-A76 could even compete with a Core i7. Actually pulling this off, even just in low-power laptops, would be an incredible achievement. But what would it practically mean for the computer industry, and why does so much ink get spent comparing products that have historically enjoyed so little overlap? Intel may have made a great deal of noise about entering the smartphone and tablet market, but it shipped few phones and its Android tablet business has been effectively finished for several years.

ARM vs. Intel: A Referendum on Intel CPU Designs

The reason these types of comparison spark so much interest, I suspect, is because enthusiasts are hungry to see if Intel (and to an extent the entire semiconductor industry) are telling the truth. For years, the basic messaging from Intel and various institutions has been that CPU performance is at a near-standstill not because of corporate shenanigans, but because the fundamental properties of silicon don’t support clock speeds much above 4.5GHz well, all of the well-known and easy tricks for boosting CPU performance have been played, and the best way to improve the performance of devices is to focus on doing so through other means. Intel, meanwhile, ran into real problems scaling its CPUs down into low power devices, and while we’ve discussed how this was much more a business decision as opposed to a fundamental lack of capability, there are still people who believe that Intel’s refusal to design high-performance microprocessors is responsible for the lack of forward progress.

If you subscribe to this argument, Apple’s leadership position in ARM silicon, rumors that it wants to remove Intel CPUs from all of its systems, and the Cortex-A76 are all proof that either Intel or some other fundamental facet of the ecosystem (like the x86 ISA) is holding back computing. In this telling, Intel’s longstanding process node leadership was essentially a desperate attempt to cover for fundamental problems in the x86 ISA that will now be exposed. Exactly what the gains are (or would be) tends to vary — some people think Intel is sitting on performance improvements, while others think it’s power consumption that could be substantially improved. But either way, the idea here is that a new approach to computing driven by ARM would prove superior to anything Intel and the x86 architecture have achieved, throwing open the floodgates to new computing opportunities.

What if ARM Builds a Better SoC?

For the sake of this discussion, I’m defining “better” to mean “a CPU that matches or exceeds customer performance requirements while matching or exceeding on battery life.” In order to really threaten Intel’s mobile market, a non-x86 CPU will need to demonstrate superiority on at least one of those two metrics.

If ARM could pull this off, it could upend the entire PC industry. Being able to smoke Intel on performance or battery life in apples-to-apples comparisons running native code would represent a true coup. It would make those of us who have insisted the laws of physics make this more-or-less impossible look extremely stupid and overturn previous studies on the absolute impact of ISA on CPU performance. But even in a scenario in which ARM has built a better SoC under Windows 10 than Intel fields, both in terms of power consumption and performance, there’s still the minor issue of backwards compatibility and emulation. Reviews of Snapdragon 835 systems have made the point extremely well; emulated 32-bit x86 performance under Windows 10 may help satisfy necessary compatibility requirements to make these systems attractive, but it doesn’t hold a candle to native x86 performance — to say nothing of using 64-bit software. The chance that ARM manages to build a CPU core so superior to an Intel x86 CPU that it can beat the latter when running emulated code is effectively nil.

As Microsoft learned with Windows RT, the software compatibility issue is a real problem that can sink an otherwise attractive product if it isn’t properly handled. And while questions of software compatibility are separate from hardware performance, as a practical question of market share we have to address both. Pushing into Intel’s core territory means offering advantages so compelling, people are willing to tolerate some software incompatibilities, which means the gains have to be pretty darn big.

I expect the overall impact of the Cortex-A76 on the existing x86 ecosystem to be modest. The performance improvements ARM is promising will improve emulation performance and boost the use of native code. It may help encourage greater adoption of the ARM variant of Windows 10, and since ARM has no real laptop market share, any gains on that front are revenue-positive for the company.


Roll Eyes

Prediction Time:     A76 will make inroads into the world of Win 10 machines.   In two more years the next one (A78?) from ARM will make even bigger inroads as it will be created specifically to do just that.   ARM moves so much quicker than Intel that Intel will never catch up to the current wave of ARM advamcements.

A76 will make even bigger motions on the Chromebook platform as a convergence of lower battery consumption and new memory types will make small light Chromebook units punch far far out of their weight class.

Intel will HAVE to react and Intel will HAVE to compete.   Expect more and better choices from Godzilla as Qualcomm has the right equipment now to "properly motivate" Intel and get them to go past their inertia point.    You can jest hear that aluminum baseball bat ring as it hits that bony Intel skull, elbows and knees     < tinggg !!! >    < tinggg !!! >    < tinggg !!! >    < tinggg !!! >    < tinggg !!! >    < tinggg !!! >    < tinggg !!! >    The sheer pain from the constant bloody beating will make Godzilla stir finally.

Microsoft will be faced with a choice, they cannot keep up with both camps by continuing to keep them separate and forcing double work on all their Windows support staff.  

Eventually one system will be dropped for the other ......

Back to top
 
« Last Edit: 07/02/18 at 15:21:00 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & issues
Reply #17 - 07/03/18 at 06:35:12
 

https://liliputing.com/2018/07/intels-9th-gen-core-chips-are-on-the-way-if-th...



Intel’s chip-naming convention has gotten kind of weird recently. So I’m not entirely sure what to make of the fact that the company included a bunch of previously unannounced processors in an official document recently.

Logically, the new Core i3 and Core i5 9000 Series chips will be branded as part of the company’s 9th-gen Intel Core processor family. But they’re also listed as “Coffee Lake S” chips, suggesting they’ll use the same architecture as some of Intel’s 8th-gen processors (or at least something similar to it.


Intel

The new CPUs showed up in an updated Intel Microcode Update Guidance document, as noted by VideoCardz and TechSpot. Here’s a list of the new 9000 series processors:

Core i3-9000
Core i3-9100
Core i5-9400
Core i5-9400T
Core i509500
Core i5-9600
Core i5-9600K

There aren’t many details available about these new chips, but the Core i3 versions all seem to be quad-core chips, while the Core i5 processors are hexa-core models. It’s unclear if or when we can expect 9000 Series Core i7 chips.

But if these are really Coffee Lake-S chips, I’m not sure what, if any, advantages the 9th-gen versions will have over their 8th-gen counterparts.

Intel’s 8th-gen processor lineup is already kind of a mess when it comes to naming. Here are some of the mobile chip families that are all branded as 8th-gen Core processors:

15 watt Kaby Lake Refresh processors
28 watt Coffee Lake-U processors with Iris Plus graphics
45-watt Coffee Lake-H processors
10nm Cannon Lake-U series processors

And that’s not even counting the company’s 8th-gen desktop chips. All of which is to say, that simply knowing the chip generation doesn’t tell you much about an Intel processor’s architecture or performance anymore. So while it’s kind of interesting to see that 9th-gen chips are on the way, I honestly have no idea what that even means at this point.



What does it mean?    Intel has issued new "safer" microcode to slow down the processing on all  their old chipsets that are still in wafer production.   The stuff they sell from this point forward will SLOW DOWN 2-5% while it does "ongoing mitigation" for the 6-8 versions of Meltdown and Specre that exist as of now.

Making a "new 9th generation" out of these "new" slowed down then paper overclocked old processors is a rare form of pure Intel marketing BS nonsense

--- certainly it isn't worth popping the cork on the champagne for Intel to celebrate anything (in my eyes, anyway).    

Calling it a "new 9th generation product" simply says how far down inside the toilet whirlpool Intel is currently rotating.

So Intel has actually quit doing some of their stupid predictive processing things that had opened up that huge raft of security issues.  To make it back up speed-wise Intel now has put a paper spec "overclock" in place that suspiciously just covers the slow down due to not using predictive processing any more --- goody goody on them.  

Small lies seem to be working better for you, huh?   

 Huh

It is all lies on paper, Intel hasn't changed ANYTHING since last year, really, except for stopping using the predictive stuff (a 5% slowdown that actually can go up to 20% for some certain select workloads) and on the top end chipsets they actually added a couple of more cores, going way past what existing cooling systems can actually handle, which means the whole shebang chokes down due to "self-preservation thermal throttling".

Heck, you will get the new slowed down processing stuff FOR FREE eventually without having to buy a new PC if you just wait for the Mickeysoft nightly update that has it in it .......    

Who would want to forgo that disastrous pleasure that that particular "break the world update" will bring ????
Back to top
 
« Last Edit: 07/22/18 at 14:41:18 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & issues
Reply #18 - 07/05/18 at 10:23:18
 

https://liliputing.com/2018/07/samsung-expects-to-develop-3ghz-or-faster-arm-...

Samsung expects to develop 3GHz or faster ARM Cortex-A76 chips for phones, tablets, Chromebooks and Win 10 laptops.



Samsung has announced that it’s working with ARM on new 7nm and 5nm FinFET process technology that will allow the company to manufacture chips based on ARM’s Cortex-A76 designs with frequencies of 3 GHz or higher.

That shouldn’t be a huge surprise — ARM pretty much said when it announced the new design that 7nm Cortex-A76 chips should be capable of 3 GHz+ speeds. But the news here is that Samsung’s Foundry is working on actually building those chips, with plans to deliver the first such chips before the end of the year.



So much for all but one of the expected REALLY BIG players (Google has not broached their new phone nor their Chromebook OP3 processor yet) ......   now we wait for Google/HTC/Rockchip to drop their A-76 news and then we will wait still more for all of the rest of the Hockey Stick Boys to go sign up for their 7nm Cortex-A76 licenses and for them to start up their very own TSMC based development for their own witch's brew custom mixtures of old and new stuff to make up their very best FASTEST TO MARKET competitor beating cauldron mixes.

Roll Eyes
Back to top
 
« Last Edit: 07/22/18 at 14:42:08 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & issues
Reply #19 - 07/06/18 at 16:03:08
 

https://www.tomshardware.com/news/china-zen-x86-processor-dryhana,37417.html

China Begins Production Of x86 Processors Based On AMD's IP



Chinese-designed "Dhyana" x86 processors based on AMD's Zen microarchitecture are beginning to surface from Chinese chip producer Hygon. The processors come as the fruit of AMD's x86 IP licensing agreements with its China-based partners and break the decades-long stranglehold on x86 held by the triumvirate of Intel, AMD and VIA Technologies. Details are also emerging that outline how AMD has managed to stay within the boundaries of the x86 licensing agreements but still allow Chinese-controlled interests to design and sell processors based on the Zen design.

AMD's official statements indicate the company does not sell its final chip designs to its China-based partners. Instead, AMD allows them to design their own processors tailored for the Chinese server market. But the China-produced Hygon "Dhyana" processors are so similar to AMD's EPYC processors that Linux kernel developers have listed vendor IDs and family series numbers as the only difference. In fact, Linux maintainers have simply ported over the EPYC support codes to the Dhyana processor and note that they have successfully run the same patches on AMD's EPYC processors, implying there is little to no differentiation between the chips.

The new chips are surfacing against the backdrop of the trade war between the US and China that could escalate quickly, likely reinforcing China's long-held opinion that a lack of native processor production could be a strategic liability. Today's wars are won with chips, and their strategic importance certainly isn't lost on those in the halls of power. In fact, the Obama administration blocked Intel from selling Xeon processors to China in 2015 over concerns the chips were fueling the country's nuclear programs, and subsequent actions by the US have largely prevented China from achieving the technical know-how and equipment to develop its own chips through acquisitions and mergers.

That makes it even more surprising that AMD has managed to establish a franchise that allows Chinese processor vendors to develop and sell x86 processors in spite of US regulations and the licensing restrictions with Intel, but now more information is coming to light about how AMD pulled off the feat.

Satisfying The x86 Cross-Licensing Agreement
AMD's announcement in 2016 that it had established a joint venture in China to develop processors was surprising, but it yielded a much needed $293 million cash infusion (in payments based on delivery dates) for the then-struggling company, which had operated at a loss for the prior six quarters. AMD is also set to reap royalties, based on unit sales, once the processors begin to ship in volume.

As part of the licensing agreement, AMD established a joint venture (JV) in China called the Tianjin Haiguang Advanced Technology Investment Co. Ltd. (THATIC) and agreed to license its x86 and SoC IP for chip development. THATIC consists of AMD and both public and private Chinese companies, including the Chinese Academy of Sciences that is heavily influenced by the Chinese government.

According to mydrivers.com, THATIC then established two companies through joint ventures: Haiguang Microelectronics Co. Ltd. (HMC) and Chengdu Haiguang Integrated Circuit Design Co., Ltd (Hygon).

But here's where things get tricky. AMD holds a 51 percent stake in HMC, while Tianjin Haiguang Holdings owns 49%. Meanwhile, AMD owns 30% of Hygon and Tianjin Haiguang Holdings owns 70 percent.

HMC owns the x86 IP and ends up producing the chips, which satisfies the AMD and Intel x86 cross-licensing agreements because the IP remains with a company owned primarily by AMD. But AMD provides the IP with the understanding that the company will use it to design its "own products specifically tailored to the needs of the Chinese server market." That requires quite a bit of maneuvering given the restrictions of AMD's x86 cross-licensing agreement with Intel.

To stay within the legal boundaries, HMC licenses the IP to Hygon, which designs the x86 chips and then sells the design back to HMC.

HMC then employs a foundry to fab the end product (likely China Foundries or TSMC). Confusingly, HMC then transfers the chips back to Hygon (the same company that designed them), which then sells the Dhyana processors.

And thus, AMD's transfer of the x86 IP stays within the legal boundaries. According to the agreement, the final products can only be sold within China's borders. That opens up a huge opportunity for AMD via royalties due to the exploding China data center market, but potentially serves a blow to Intel due to the Chinese governments' influence: China has invested heavily in native chip producers through its Made In China 2025 initiative and also offers incentives and other measures to prop up domestic chip production.

China Continues To Invest
It's unclear what the impact of AMD's licensing agreement will have on the long-term outlook for China's chip development ambitions. China is also heavily involved in other chip-producing ventures, such as Zhaoxin Semiconductor, which is working to produce x86 chips through a partnership with VIA. The combination of these efforts, among others, could ultimately provide the country with the independence it desires from US interventions, and possibly alter the long-term processor market in the process.


WHAT DOES IT MEAN FOR INTEL?

Intel had hopes of throttling AMD back to "only taking 20% of their market share" away in the second half of 2018.

That goes out the window now as you have 3-4 Chinese government backed companies LICENSED to use AMD Zen and Via x86 tech in a mix or match fashion with MIPS and IBM and ARM technologies.   This is legal and above board.

Hygon must make the chips and sell the chips inside China.  This does not limit secondary or tertiary vendors (motherboard or systems builders) from reselling the final end product through a supply chain that ends possibly outside of China.

What to be watchful for is what is NOT above board --- China's very capable industrial espionage functions might be lifitng tech from everybody and then producing it as a "subsidiary development of combining Zen and IBM and Via technologies".   Micron as a memory maker has already been pillaged by China so their new memory stuff is coming soon from the Far East now too.  

Plus China is BLAZING FAST, they can steal something and have it in full production as their "home designed subsidiary development tech" before the original company can finish up their lengthy required in country testing and then arrange for a production slot at TSMC.    

As far as legalities go, obviously Chinese company XYZ had also developed the tech on a parallel pathway using old tech they had licensed from AMD, VIA, IBM, ARM, MIPS, (you get the idea, Chinese companies hold mfg and development rights to a whole lot of old American companies old tech nowadays).

In cases like AMD, some older tech was licensed outright -- the fact that some new features fresh from brand new industrial espionage from vendors B and C got mixed into the pie is "regrettable".    

When the total service life of a processor is only 2-3 years, it can easily take 4 times that long to reach some sort of minimal court settlement based off litigation in the Far East.   And that is if it doesn't wind up as any form of shared judgement due to fault existing on both sides.

CHINA IS NOW A MULTI-FOUNDRY PC LEVEL & SERVER LEVEL CHIP SUPPLIER starting today.

Look to see a whole lot of PC and laptop action coming out of China sometime out in the future ..... dirt cheap just like the little Android TV boxes are today.
Back to top
 
« Last Edit: 07/13/18 at 06:57:47 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
The Empire Strikes Back
Reply #20 - 07/12/18 at 18:11:43
 

https://liliputing.com/2018/07/intels-new-xeon-e-series-chips-are-for-entry-l...

First Moves from the Empire side of things, things done after the removal of the Evil Emperor.





Intel is launching a new line of Xeon E processors designed for desktop workstation computers.

The company says the new chips offer up to a 36 percent performance boost over previous-gen low end server chips ...... which is probably a more appropriate spec for companies that might only update their workstations about twice a decade.

The company is offering 10 different Xeon E chips, ranging from the Xeon E-2124 processor with 4 CPU cores, 4 threads, no integrated graphics and 8MB of cache to the Xeon E-2186G with 6 cores, 12 threads, 12MB of cache, and Intel UHD 630 graphics.

All of the new chips have 40 PCI Express 3.0 lanes, support for up to 64GB of DDR4-2666 RAM, and they all fit motherboards with an LGA 1151 socket.

Other features include support for Thunderbolt 3.0, USB 3.1 Gen 2, Intel Optane memory, and Gigabit Ethernet. TDPs for the chips range from 71W to 95W, depending on the processor.

Dell, HP, Lenovo, and other PC manufacturers are expected to offer workstations with the new Xeon E chips soon
.

Yes, you really do see Chipzilla moving preemptively to block the easy road that Hygon would have taken to steal lots of market share from Intel over in the China Zone using the bone stock AMD tech Hygon has already used to produce whole new runs of finished chipsets.

There is no doubt that this move is what Intel was referring to as "stopping AMD from gaining large amounts of server market share past 15-20%".   By adding two extra CPU cores and leveraging a large chunk of faster buffer memory Intel really has bumped their performance by 36% on the top end units as compared to existing server processors.  

Take the 36% as a temporary paper gain, thermal throttling will steal that back away from the end user and potentially a lot more besides.   Consumer level cooling systems cannot handle these big Intel bruisers -- you pay a lot for potential performance you will only see in the first minutes after starting to work.   Then the Intel nanny chips thermal throttle the beast down to give you less than a two year old (two cores smaller) chipset could give you.

IT IS MY OPINION that Intel is functionally giving up on the low end of the laptop/desktop computing market and is ceding it to ARM as all their competitive moves seem to be in the Xenon class of server processors which are very rapidly becoming Intel's standard "gaming" and "business" level processors.

Now for the first counter moves from AMD and Hygon and the Google Chromebook people.


==================================================


https://liliputing.com/2018/07/lilbits-319-chrome-gets-more-secure-uses-more-...

Chrome gets more secure, uses more RAM



Google’s Chrome web browser has always been something of a memory hog, which is to be expected from an application that runs a new sandboxed process for each browser tab.  

This < 10% relative increase in the memory used by ChromeOS pales compared to the amount of memory used by ANY Microsoft OS system on ANY Intel processor which always uses twice as much memory and twice as much processor power to run at relatively similar speeds ...... simply because it is fat and porky and slow (i.e. it is Windows 10).

Now Google has introduced something called “site isolation,” which means that not only does each tab have its own process, but each sub-process can also now only render documents from a single site, even if there’s a cross-site iFrame in a page.

Such site isolation should prevent malware from leveraging the Spectre vulnerability affecting modern processors to steal your data but it comes at a cost of a 10-13 percent increase in memory usage.


Point here --- Google Chrome has actually DONE SOMETHING REAL to address the two forms of Spectre and the six various forms of Meltdown.   Google's Chrome fixes are REAL and they all are turned on by default and they all are fully active at this point in time.

At this point Wintel has only TALKED a very partial sort of game and has only put out some partial changes that ARE NOT TURNED ON BY DEFAULT, and since they remain turned off may as well not exist in the real world.   Their users are all exposed still to getting their data stolen.

ChromeOS is still so light and fast it still kicks Wintel's browsers butt speed-wise even with all the extra sandboxing and all single site/tab memory isolation requirements in place and working.

Chromebooks are still growing and are taking more and more laptop share away from Wintel.

https://www.statista.com/statistics/749890/worldwide-chromebook-unit-shipments/     Look at the matching Windows PC unit loss charts along the bottom edge of this chart.

We still see Wintel moving all their improvements over into the Xenon chipset family simply to get enough cores and enough brute processing power to get out of its own way and show some positive motion.    We see Xenons slowly becoming i9s and i7 class as these bulkier more expensive items gradually become Intel mainline processor chipsets.

This functional doubling/tripling of core counts and the associated HUGE memory amount requirements (compared to ChromeOS) makes it a VERY EXPENSIVE counter move by Wintel .......      

Expensive or not, it is the only move Intel has on tap at the moment, so they are making it -- Intel is doing this to response to ALL their various competitors at all the different levels in their businesses.

Roll Eyes

I guess mebbe Intel got all dizzy, spinning around in that tight tight little circle down inside that toilet bowl whirlpool .....
Back to top
 
« Last Edit: 08/07/18 at 06:33:05 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & issues
Reply #21 - 07/16/18 at 20:54:37
 

https://www.tomshardware.com/news/intel-micron-3d-xpoint-imft,37461.html



Wow, it is an original Optane 3d XPoint board, suitable for hitting a PCIe card slot on a specialty rack space motherboard that was constructed specifically to use it as a buffer memory for a large platter hard drive.    Large Samsung SSD drives were less expensive and ran a good bit faster in real use as you could put your whole hard drive on the bootable Samsung SSD for the same money and the original Intel Optane drive was always restricted to buffer functions only.

Intel has been bundling these boards with the specialty rackspace motherboards that mount them as a "special deal" as sold by their favorite board vendors, essentially giving the Optane boards away for free lately since they have a warehouse full of them.    

Intel has a whole generation of rackspace motherboards with Intel CPUs and matching Optane boards jest a sitting in a warehouse that have been totally overcome by events, so ya wanna buy one of these anyway?   Cheap?   Intel will give you the Optane stuff for free ......




This is the DDR4 version of the same sort of memory, once again requiring a specialty board from an Intel favorite board vendor.   Very few sample sticks or boards were ever built as the cost was way too high and the Optane stick memory was too slow again --- Samsung standard memory sticks were a lot better deal price/performance wise.

Tom's Hardware covers the pending death of Optane memory in the blue quote below as the final stages of the Micron Intel divorce become known.    Be aware that Optane won't completely die and disappear until the warehouses get empty, sometimes in late 2019-2020.

What will happen is that the other forms of fast non-volitile memory (faster and cheaper) are what will end Optane in reality and will be what sweeps it off the table starting late next year.

Micron is done with Optane, sees no future for it as it isn't speed or cost effective compared to the new SD Association Level 7 and new proposed Level 8 SD and MicroSD standards that all the memory companies are supporting as of now (Micron included).

So, Intel is once again the only one sitting in the leaky 14nm Optane row boat at the moment, and Intel has simply lost track of their oars again and is just floating along until the warehouses finally all get emptied out.


XPoint Shakeup: Intel and Micron to Cease Joint Development of 3D XPoint Next Year
by Paul Alcorn July 16, 2018 at 4:15 PM

Intel and Micron have announced they have updated the terms of their 3D XPoint joint development partnership and will cease joint development after the second generation of 3D XPoint is completed in the first half of 2019.

The two companies also recently announced they would cease joint development of NAND through their Intel-Micron Flash Technologies (IMFT) partnership after the third generation of flash technology, but the storied IMFT franchise will apparently soldier on through shared production facilities.

Intel and Micron will continue to develop new generations of 3D XPoint independently "in order to optimize the technology for their respective product and business needs."   This means Intel representatives confirmed to us that Intel will still produce 3D XPoint out of the jointly-operated IMFT fab in Lehi, Utah even after the companies develop their respective third-generation products, but Intel still has the option to produce the memory at other facilities if needed.

Intel and Micron announced 3D XPoint in July 2015. The companies designed the new memory to bridge the performance gap between NAND and DRAM, and because of 3D XPoint's persistence (it retains data after power is removed), it can serve in both memory and storage roles.

Micron announced its 3D XPoint-based QuantX products in 2016, and we even tracked down the finer details of the design, but the SSDs did not make it to market. As a result, Intel is the only company selling 3D XPoint-based products and Micron has relied on Intel to purchase its excess 3D XPoint production inventory.

However, during a recent earnings call, Micron CEO Sanjay Mehrotra revealed that 3D XPoint sales to Intel were flagging, thus incurring under-utilization charges that impacted Micron's bottom line. Micron even indicated that it was possible that the company wouldn't sell any 3D XPoint to Intel in the future. As such, Mehrotra announced that Micron would re-negotiate the terms of future 3D XPoint development with Intel. That means the split in product development is likely the outcome of the renegotiation.

Sales of 3D XPoint have obviously been under expectations. Intel has brought the new memory to market in various SSD form factors, but pricing has been a concern as prices for traditional flash-based SSDs have continued to plummet. Intel has even taken to offering "Core+" processors that come bundled with 3D XPoint drives in an apparent effort to spur sales.

Analysts have long predicted that the debut of Optane DC Persistent Memory DIMMs, which bring the speedy memory to the DIMM form factor to enable explosive memory capacity increases, would mark the true turning point for 3D XPoint adoption. Unfortunately, Intel's 3D XPoint DIMMs have been plagued by delays, which ultimately has led to lower-than-expected sales.

Intel is currently ramping production of its long-overdue DIMMs and claims that several hyperscalers and cloud service providers are committed to deploying its products. Micron says that it will bring its first 3D XPoint-based products to market by the end of 2019, with meaningful revenue occurring in 2020.

Both Intel and Micron have sunk a tremendous amount of R&D into the skunkworks-class project over the course of a decade, so it is unlikely that either will cease development and production. Intel and Micron have repeatedly stated that 3D XPoint is an inherently scalable design: the companies can either add more layers, shrink lithography, or store more bits per cell to boost capacity and performance. That leaves plenty of room for improvements in the future. Given the announced timeline for the second-gen products in early 2019, we could see faster, cheaper, and more capacious 3D XPoint products soon
.

This last paragraph ignores the fact that other forms of persistent memory from folks like Samsung are EATING OPTANE'S LUNCH RIGHT NOW for both performance and price with the standard SSD and the standard DIMM style memory being run at 10nm and 12nm at Samsung and at TSMC.  

When 7nm persistent SSD and DIMM memory products from Samsung and others gets into the bulk distribution channels Intel hopes it will have given most of the existing stocks of 14nm Optane sold or given away by then as their aging Optane products will have become relatively undesirable even as buffer memory at that point in time.

Micron wants to get away from Intel and is taking all the needed steps to firm up the divorce settlement.

Undecided

Do not buy an Optane board -- it is EXPENSIVE poorer performance memory for your computing dollar.
Back to top
 
« Last Edit: 08/07/18 at 10:32:54 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & issues
Reply #22 - 07/17/18 at 15:33:18
 

https://liliputing.com/2018/07/intel-and-micron-announce-2nd-gen-3d-xpoint-me...

Roll Eyes

So, Intel finally releases a press pack spinning things their way, saying Gen 2 Optane is coming "in a few months" and will look like this ......



These are the newest Intel provided PR images using some various standard components (a video card and a SSD hard drive that were spray painted black) since nothing depicted here has even been final designed at the Generation 2 level by Micron yet and likely never will be as Micron wants Intel to finish paying for what they had contracted to have built at Gen 1 last year before Micron spends another second or another penny on doing any more Optane anything for Intel.

Visual items to carry away in your thoughts ...... Optane runs hot and converts a lot of wall socket power into HEAT that has to be removed by big heat sinks and buzzing channeled fan cooling actions.

Not suitable for a laptop at all, in other words .......


Here is the rest of the stuff that flows from the re-tuned Intel PR release .......   (remember Intel exaggerates and lies a lot lately)

Solid state drives tend to be faster than hard drives. And RAM (random access memory) tends to be faster still… but it’s “volatile” memory, which means that any data stored in RAM will disappear when you reboot or turn off a computer.

A few years ago Intel and Micron teamed up to develop a new type of non-volatile memory that’s almost as fast as RAM, allowing it to be used as storage, memory, or both.

Intel has brought a number of 3D Xpoint products to market under its Intel Optane brand. So far most have basically been speedy solid state storage drives, but the company recently introduced its first 3D XPoint memory DIMMs for use in data centers.

Now Intel and Micron have announced that they expect to complete development of 2nd-gen 3D XPoint technology in the first half of 2019. But that’s effectively going to be the end of the partnership between the two companies.

There’s no word on what kind of performance or features to expect from 2nd-gen 3D XPoint technology. We’ll probably find out more in the coming months as the companies get ready to launch new products based on their work together.

In the same press release where the companies announce the impending completion of the 2nd-gen 3D XPoint technology, Intel and Micron also say that they have no plans to work together on a 3rd-gen:

Technology development beyond the second generation of 3D XPoint technology will be pursued independently by the two companies in order to optimize the technology for their respective product and business needs.

While that certainly doesn’t spell the end of 3D XPoint, it does suggest that Intel and Micron have different plans for the technology moving forward.

For now, the companies plan to continue manufacturing memory at an Intel-Micron Flash Technologies facility in Lehi, Utah




===================================================



https://www.trustedreviews.com/reviews/samsung-950-pro-m-2-256mb-and-512mb-ssd/

OK, this is what killed Optane starting last year, and it looks like it will complete the job this year.   Key thing to remember, no fans -- it doesn't get hot and suck power just sitting there like Optane does.


These are the Samsung 256 and 512 gig versions of "solder on" phone motherboard memory chips which are shipping now in current top line phones.



Also new to these plug in Samsung SSD drives is the use of the NVMe standard, which we first saw with the lightning-fast Intel SSD 750-series drives.  This is a new interface protocol specifically designed for laptop SSDs – rather than previous standards which were developed for hard drives – and brings with it a boost in speed right across the board.



Samsung 950 Pro M.2 256MB and 512MB SSD

Again, though, you’ll want to confirm that your motherboard/laptop offers support for all these new features, particularly with NVMe. Without BIOS/UEFI support, the drive may continue to work but it may not be bootable.

The above features combine with Samsung’s latest 3D V-NAND memory chips. Here, the transistors that make up the memory are stacked vertically as well as horizontally, which means they’re far more dense than conventional NAND.


The result is Samsung claims that the 512GB drive chip can read at up to 2,500MB/sec and write at 1,500MB/sec, while the 256GB drive can read at 2,200MB/sec and write at 900MB/sec. What’s more, the drive chips have a total write life of 400TB and 200TB respectively, which amounts to completely filling and rewriting the drive more than 700 times.


As such, Samsung has been able to include an impressive five-year warranty with these drives, and it rates them as having a mean time between failure of 1.5million hours.


Read more at https://www.trustedreviews.com/reviews/samsung-950-pro-m-2-256mb-and-512mb-ss...



AND I REPEAT FOR EMPHASIS --- DO NOT BUY AN OPTANE ANYTHING, EVER.



Follow on next day stock market responses to these new Intel PR revelations is Samsung UP,  Micron UP,  Intel DOWN.    The stock market rewards Micron for ditching Intel and for cutting off the Optane bleeding .....   Samsung gets rewarded for cutting edge price/throughput/durability/low heat & the best battery life.


Back to top
 
« Last Edit: 07/18/18 at 07:30:19 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & issues
Reply #23 - 07/18/18 at 15:06:48
 

This comes under the heading of "Intel sez" followed up by Apple actually building some of them and finding out the hard way.



https://youtu.be/Dx8J125s4cg     

Yep, it is an Apple guy video producer complaining in a video (very well done video, BTW)  --- click on it to watch it as it is the whole refrigerator freezer show and tell.

"Intel sez" .....     Here is your spec sheet, you can trust it.   Don't worry about having to do any of that pre-release testing stuff, it just makes your hair get gray early ......


"Intel sez" .....     Which is even more dubious than "Rossi sez" as Rossi can occasionally surprise you by surpassing his own brags on some select occasional items.   And it turns out Rossi told you the real literal truth, in as much as anyone understood it at the time.

https://liliputing.com/2018/07/lilbits-320-apples-new-macbook-pro-is-fast-exc...



https://youtu.be/Dx8J125s4cg    it is a video, click on it to watch it







Lilbits 320: Apple’s new MacBook Pro is fast… except when it’s not
07/18/2018 at 5:41 PM by Brad Linder

Apple launched its most powerful MacBook Pro laptops to date last week, including 15 inch models with support for up an Intel Core i9 processor and AMD Radeon Pro discrete graphics.

Unsurprisingly, a top-of-the-line model scores really, really well in benchmarks, and Laptop Magazine reports it has one of the fastest SSDs shipping in any laptop.

But YouTube Dave Lee discovered that when you push the laptop up towards its normal working loads… it slows down considerably. It looks like the MacBook Pro chassis might not be able to keep the motherboard cool enough for the Core i9 processor to run at listed normal rated speed for any extended period of real use.

So while some tasks are super-fast, others (like heavy-duty video rendering jobs) can take longer on the new 15" MacBook Pro than on older (fewer cores) models from year before last.    

Whoops?

On the bright side, the laptop is super fast if you stick it in a freezer while rendering videos.

So that’s always an option.



==================================================


So, if you thought that one was a good one, here is a brand new breaking news one that might even wind up being an even better "chicken choker" than a 15" MacBook Pro .....

https://liliputing.com/2018/07/hp-updates-its-z-line-of-entry-level-workstati...

HP updates its Z line of entry-level workstations with Xeon E options (and more)
07/18/2018 at 9:00 AM by Brad Linder

HP is updating its line of entry-level workstation computes with four new models, including the HP Z2 Tower, HP Z2 Small Form Factor, and HP Z2 Mini, along with the EliteDesk 800 Workstation Edition.

The smallest of the bunch is the HP Z2 Mini G4, which is an update to the Z2 Mini Workstation that launched a few years ago.

The new model is the same 8.5″ x 8.5″ x 2.3″ size as its predecessor, but under the hood it packs a lot more power, with up to NVIDIA Quadro P1000 graphics, up to an Intel Xeon E-2176G processor, and up to 32GB of DDR4-2400 memory.




This is actually a little hover craft with four corner lifting air jets .......

Grin

The little computer has room for a 2.5 inch drive and an M.2 NVME 2280 SSD and features a USB 3.1 Gen-2 Type-C port, four USB 3.0 ports, Ethernet, DisplayPort, and a headset jack and two SODIMM slots.

The HP Z2 Mini G4 supports 802.11ac WiFi and Bluetooth 5.0 and in comes with a bunch of processor and graphics options including Intel Pentium Gold through Xeon E chips and a choice of AMD or NVIDIA graphics.

Prices start at $799, and HP is positioning the Z2 Mini workstation as not only small enough to put on your desk without taking up a lot of space, but also as a system that can be mounted to the back of a monitor or set up in a rack.

Undecided      Tongue      Shocked

I mean if you are gonna REALLY REALLY choke a baby chicken,  putting an overclocked rackspace derivative Xenon E2176G chipset, 32 gigs of systems RAM and a NVIDIA Quadro P1000 graphics card all inside a Chromebox sized case with very little air flow and absolutely no effective heat sink fins built into the PLASTIC case top .......       Roll Eyes    

Love that toolbox sized "laptop floor charger" ya got over there, guys  ..... but what's with the two clear plastic water lines running beside that really thick power cord going up to the little box on the desktop ???   And why is that big boxy power supply so durn noisy ????


Really ????    HP guys, do you really even read your own press releases ????

Or did Intel build and release this particular press release for you guys as a "time saving favor" to you ????


Apple is jealous that HP has done gone and one-upped them again on the overheated front .....

...... this time fer choking very small baby sized chickens even worse than Chick-fil-A ever did
Back to top
 
« Last Edit: 07/24/18 at 12:10:52 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & issues
Reply #24 - 07/21/18 at 23:58:49
 

https://9to5mac.com/2018/07/21/intel-power-gadget-mac-removed/

Mac Pro 15 users frothing over Intel Core i9 Chip performance

This is the original story, which gets into the disaster which is Intel's Core i9 chipset, a disaster composed of adding 2 extra cores to an already expensive premium chip, cores that are actually counter productive in that they jerk up the whole chips thermal curve up higher to the point the whole chip throttles to perform WORSE than before the extra cores were added.

Intel diagnostic tools were used to discover the details of these traits, now those diagnostic tools have been yanked by Intel and the diagnostic tools were reworked so they show something else for results.   Trouble for Intel is that other non-Intel diagnostic tools do exist that now no longer are in agreement with what the tweeked Intel tools now show.

Intel lies, gets caught at it and lies again ...... there is a pattern of deception here that is only now becoming bleedingly obvious to the Mac crowd.

The download link for the Intel Power Gadget, which reports info like Mac CPU temperature and current clock speed, has been conspicuously removed from the Intel website today. There’s no explanation on the page as to why the company suddenly removed the download, although it sure is convenient timing. The utility has been used by many tech reviewers to highlight possible thermal problems with the 6-core 2018 MacBook Pros.

Perhaps Apple or Intel have discovered inaccuracies in the tool’s reporting and is now in the process of updating it.

The latest version of the app, Intel Power Gadget 3.5.2, was available for download for free up to last night. Various posters on Reddit are uploading mirrored copies for people to download now, given Intel’s official download is currently unavailable.

The latest-gen MacBook Pros retain the same hardware chassis design but feature new, power hungrier, quad-core and six-core CPUs. Particularly on the highest-end model, the 15-inch MacBook Pro with Core i9 chip, a number of reviewers have noted severe performance throttling, with the Power Gadget reporting the processors are dropping below their base clock speed with longer-running tasks like video export.

YouTuber Dave Lee brought this into the spotlight, demoing that the 2018 MacBook Pro was actually slower than the last-gen model at an Adobe Premier video export — seemingly because the new i9 chip kept having to be significantly throttled to maintain internal temperatures.

In our own testing, (9to5mac.com) we found that disabling two of the cores actually resulted in faster export speeds in Final Cut Pro than if all six-cores were active.

One of the things that has come out of this whole debacle is that the Intel Power Gadget can report CPU frequencies as low as 800Mhz. However, this does not necessarily mean that the CPU is being throttled — it likely means the CPU is idling whilst other components like a hardware decoder are operating. It is possible that Intel is now updating its tool to better explain the dips in frequency.

At the time of writing, only the Mac version of the Intel Power Gadget is unavailable. Downloads for Windows and Linux are still offered.


Intel's Power Gadget Utility is back up again, but shows far less throttling and no longer aligns with other tools nor with observed reality all that well.   It is becoming clear that Intel's "add a new pair of cores" response to AMD competition isn't working out flawlessly for Intel as Cooling becomes Critical to that desperate Intel competitive response and the existing crop of laptop products simply lack the necessary cooling abilities to handle the much hotter running Intel Core i9 and up chipsets.

It has been discovered that Intel is also playing games with interleaving larger amounts of time for "idling for hardware decoding" into both their firmware and into the Gadget Utility test tool. These "delay items" are not counted against the chipset performance by the existing crop of processor utilities.   Using these minor dirty tricks in the Gadget Utility test tool that simply give the ever hotter Intel chipsets more time to cool down while disguising the root cause instead of pointing it out clearly.

This goes to support the thought that the new "improved" Intel chips really do actually perform about the same as the old chips from 2-4 years ago, except for the very newest ones with the extra cores which can actually perform WORSE due to excessive overheating and throttling.



===================================================



Apple finally speaks ....... four days later.

https://liliputing.com/2018/07/apple-acknowledges-macbook-pro-cpu-throttling-...

Apple acknowledges MacBook Pro CPU throttling, promises update to fix it
07/24/2018 at 2:40 PM by Brad Linder

Apple’s most expensive new MacBook pro laptops are available with up to a 6-core Intel Core i9 processor and AMD Radeon Pro graphics. Unfortunately early testers have discovered that the laptop can overheat very quickly, causing the speed to drop so low that previous-gen MacBook Pro models are actually faster.

That’s kind of disappointing for a laptop with a price tag that can easily top $3,000.

But Apple says it’s aware of the problem… and the company is releasing a software update that the company says should resolve the issue.

According to Apple, there’s “a missing digital key in the firmware that impacts the thermal management system,” which could cause CPU clock speeds to take a nose dive under heavy load.

In other words, when the computer gets hot, the CPU will slow down until things cool off… which means that you could theoretically render videos or perform other CPU-intensive tasks more quickly on an older laptop with a less powerful processor.

Ars Technica notes that it was pretty easy to confirm this behavior: Cinebench, Final Cut, Adobe Premiere, Geekbench, and other programs all ran faster on a 2018 MacBook Pro with a Core i7 CPU than on a model with a Core i9 chip.

Today Apple is rolling out a software update that should correct things so that the Core i9 models do indeed run faster.

Apple says the bug fix is part of the macOS High Sierra 10.13.6 Supplemental Update that’s rolling out to users today.

It seems odd that Apple shipped a high-end computer with a bug so obvious that it was quickly discovered by some of the first people to use the laptop. But it’s good to know that Apple believes it was a software bug rather than a hardware issue… which would be a lot tougher to fix.

That said, I’d wait until we start to see post-update benchmarks before buying a 15 inch MacBook Pro with a Core i9 processor.



Why this last sentence ????   Teardowns of the MackBook Pro series of laptops shows a totally inadequate CPU cooling system is what is installed in the MackBook Pro laptops, a CPU cooling system that is not adequate really for anything past the M series chipsets (which really don't require a cooler but get one anyway in the MackBook Pro series of laptops).    

The higher energy AMD Radeon Pro graphics chipset gets no extra cooling at all, and that too is a mistake on a $3000 gaming capable laptop like this one.

Somebody at Apple lost track of what the heck they were doing, looks like.    Look for a refund or a unit replacement call back to be announced after this interim software fix fails to work right.



===================================================



Apple fixed what they could, found settings that were not correct as shipped and tuned those.

Apple simply did the homework that they were too slack to do before they shipped the stuff and Apple got called on it.

Here are the results ----- it is better, but not nearly as good as a Windows Gaming unit, or even a standard Windows unit with the same CPU, amount of memory etc.    

Apple simply did not put enough cooling into the laptops (being too concerned about looks to make it as fat as it needed to be).  


https://youtu.be/UTguywiC9aw     it is a video, click on it


Intel is simply slamming extra cores onto already hot running chipsets and taking them completely off the edge of the world thermally speaking, especially in the Apple products.

COMPLETE REDESIGNS of laptop and cooling system are needed to use this 8 Core i9 stuff.

Windows machines have done some of this, Apple products have not.



Back to top
 
« Last Edit: 08/04/18 at 12:16:53 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & issues
Reply #25 - 08/04/18 at 10:58:03
 

https://www.tomshardware.com/news/amd-fenghuang-ryzen-vega-soc,37551.html

The mysterious AMD Fenghuang is a complete system on chip (SoC) powerhouse made up of a quad-core Ryzen processor paired with 24 Vega Compute Units and 8GB of GDDR5 memory on a single chip.

Kaby Lake-G ushered in a new class of processors for Intel, with all the progress items coming from AMD.   Now that AMD's short marriage with Intel has ended, the red chipmaker is free to branch out and do its own thing.   Jack Huynh, AMD's corporate vice president of its semi-custom business unit, announced today via the company's blog that it has created a new SoC for Chinese electronics manufacturer Zhongshan Subor.

The semi-custom SoC, codenamed Fenghuang, will power Zhongshan Subor's upcoming gaming console and PC. The Fenghuang chip harnesses the power of AMD's Zen processor and Vega graphics processing unit architectures. The SoC features a Ryzen processor with four cores and eight threads running at 3GHz and 24 Vega Compute Units operating at 1.3GHz. An Accelerated Processing Unit (APU) might be the first thing that comes to mind; however, the Fenghuang SoC is quite the opposite. APUs don't come with their own graphics memory, and the Fenghuang chip has 8GB of high-performance GDDR5 memory at its disposal. Jack Huynh also confirmed that the SoC will support the chipmaker's existing and next-generation technologies, such as AMD FreeSync, Radeon Software Adrenalin Edition software and Rapid Packed Math.

Zhongshan Subor demonstrated its gaming PC at its booth at ChinaJoy, one of the largest gaming and digital entertainment expos in Asia, held in Shanghai. The Chinese manufacturer stated that it plans to release the gaming PC in late August. The SUBOR gaming console, with the same specifications as the company's gaming PC and a customized operating system will be available by the end of this year.


AMD and Intel are at the approximately 50-50% mark on new unit PC sales in the USA at this particular point in time.

AMD did something sorta new this week, something for desktop and console units that was requested by a China console firm.  They put together a single chip SoC format that had ALL the requested functions on the chip itself, including a largish chunk of fast access on-die operating systems memory, 32 gigs worth of it sitting up on an on-chip PoP package.  

This makes up a complete SoC package for the console units that can also be used in a laptop or a desktop .....

This is ground breaking stuff for desktop uses, utilizing pure cell phone type low cost ideas in PC space by clear intent.  

The AMD SoC at 12nm currently only requires a large flat style copper heat sink so it can fit and work in thin laptop formats as well.    Once it shrinks to 7nm it will work even better at a lower required heat dissipation level.

Expect low end laptop uses and small desktop uses to proliferate like crazy out of China (which is starting now even as we speak).    China based Zhongshan Subor has licensed the build rights for this Fenghuang style processor.   We do not know if Zhongshan Subor has any design rights arranged with AMD at this point in time .....

Intel has only had one reply to all their many challengers at this point in time and that is MORE and MORE and MORE bigger hotter running cores ---- this thought is failing big-time on the overkill & over expensive side of the equation.  

Thermal throttling is the bugger bear that is biting Apple and Intel in the arse at this time -- getting rid of the heat is the challenge that Apple has failed to address.

Intel understands they are going to lose all lower end market share due to intense new competition from ARM and AMD ..... but Intel is bound and determined to hang on to the upper end (the very most profitable) segment of the PC processor market.   Intel losses in rack space are beginning to seriously mount up STARTING NOW as the Chinese Dhyana rack space CPUs are shipping inside China as we speak.  

China is the largest rack space growth market too, and Intel holds almost no new rack space installation sales inside China proper at this point in time.  

China as a government and as a people strongly prefer to buy a Chinese built chipset due to Trump's recent trade war attacks and his two year's worth of various Intel Xenon export restrictions.

So, Intel is shedding market share right now like the Wicked Witch of the West did once the mop water hit her in the face.



MELTING .......  melting ......  (groan) ......  melting        
My Intel Inside worldwide market share is melting .....





http://fortune.com/2018/06/18/amd-stock-price-intel/

Gus Richard, an analyst at Northland Capital Markets, on Monday downgraded his rating on Intel’s stock to “underperform” with a price target about 20% below Intel’s closing price on Friday. Increased competition from AMD, greater use of graphics chips in many data centers (for performing AI tasks), and a diminishing ability by Intel to unveil chip manufacturing advances ahead of rivals are all hurting, Richard noted.

AMD (AMD, -1.70%) has lately been updating its revived chip line with second generation improvements to the Ryzen chip for PCs and Epyc chip for servers.

“The performance gap between Intel and AMD has narrowed substantially,” according to Richard. “AMD is now well positioned to compete in desktop and notebook markets. While AMD’s parts may sell at a discount to Intel’s the discount is declining and the price points are increasing.”

Intel said it was ready for the competition. “While we are prepared for a more competitive environment as we move through 2018, we’ve already factored that into our financial forecast and we’re in a great position to compete,” Intel said in a statement. “We remain very confident in our products, our roadmap and our competitive position.” The company pointed to its Xeon processor Scalable family and upcoming new Intel Optane DC persistent memory and storage technology as examples.


This last paragraph once again shows Intel REALLY wants the business world to remain completely oblivious to the fact that Micron has dumped them as an Optane partner and the Core i9 and Xenon Scalable Chipsets (same stuff actually) are really running head on into some very hard thermal throttling issues.    

Those last two cores that Intel tacked on were two cores too many -- older lower core count processors from two years ago can do the same real workload items quicker unless the new Core i9 processor equipped laptop is literally put inside a freezer to keep the works cool enough for all the cores to function at Intel's intended specified rates.



====================================================



https://www.zdnet.com/article/computex-2018-amd-previews-32-core-2nd-gen-thre...

As Intel struggles to find cooling solutions for their newest Core i9 and other rack space chipsets that they now call "Gamer" chipsets, AMD has done what they said they would do at Computex inside the 3 months time span they said they would do it inside.

32 cores, 64 threads, it goes into existing dual socket standard AMD thread ripper motherboards with existing standard cooling options fully workable --- a rational established gamer technology that is naturally unlocked and completely over-clockable.

It is here.   It is real.   For any test you care to mention it completely trashes Intel's very best 24 core  "promised for production but only available as samples at this time" Core i9 / Epic / Xenon chipsets.

The 32 core AMD chipset is 53% faster than Intel's 24 core best and it does not have the severe throttling issues that haunt the power hungry hot running 24 core Intel unit.

AMD has added some sweet sweet frosting on the top of their cake, the CPU itself costs $400 less for each processor, that is $800 per pair less than existing Intel offerings that do not perform as well and make more heat and throttle a lot more .....  and Intel will require you to buy new sockets and new motherboards and some expensive fancy memory stuff that AMD does not require to get up towards that better speed levels.



===================================================



https://www.zdnet.com/article/microsofts-got-a-new-plan-for-managing-windows-...

I have been saying for the last longest time Microsoft was going to a subscription basis for you to use their Windows 10 software.

This starts this month in the Business world first, as a new Microsoft run MONTHLY subscription system for your PC hardware complete with all consumer/business MS OS and software products.

MS is branching out, it seems ........  and they are really going to own your PC now legally  .....
Back to top
 
« Last Edit: 08/07/18 at 15:14:10 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
justin_o_guy2
Serious Thumper
*****
Offline

What happened?

Posts: 55279
East Texas, 1/2 dallas/la.
Re: 2018 -- new Intel failures & issues
Reply #26 - 08/06/18 at 16:52:44
 
Mickey should just start leasing computers out.
Back to top
 
 

The people never give up their liberties but under some delusion.- Edmund Burke.
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & issues
Reply #27 - 08/07/18 at 03:34:43
 

Yep, Justin --- MS will in essence lease the hardware out along with the software, retaining ownership of both, retaining legal rights to go in and alter both at will at night as the nightly updates roll.   For whatever reason, be it good or bad in the eyes of the user.

The rub becomes cost, add those monthly charges up and you are really just buying your computer over and over and over again as the years roll by ......

Google has a much better deal for overall cost .....    Mickey needs to figure out how to get their total ownership cost DOWN a bunch ASAP as Google is actively moving in on business now, gaining business acceptance and increasing their market share significantly.

We are seeing Mickey's first reaction moves to the Chromebook Crostini System for Linux and what Crostini is doing to make both Linux apps and Windows apps able to be loaded on to a standard Chromebook in a painless fashion.



===================================================



I know I have been telling you guys for a while now that all Linux Apps are all coming to Chromebooks soon.   This has happened already for 18 Chromebooks now in the Developer mode, but rumors are now spreading that the last major bugs to Crostini have been worked out and on one of the very next general Chrome OS release numbers this feature will go live across the board as standard, just like the Android app support has done.  

This means Chrome 69 or Chrome 70 will have this Linux ability naturally, which is sensible since Google has already put in full isolation Sandboxing into Chrome 68 which opens the door wide for the Linux (and for those nasty funky Windows apps) to safely run inside their own little sheltered worlds inside your modern Chromebook.

This latest move by Google has tightened the screws on MS big time, and is thought to be prompting MS to only be willing to release their newest apps when combined with a totally locked down hardware and OS system set up.

Business will quickly disabuse MS that they are calling the shots in this area as Chrome/Linux/Wine is willing to load your old version Windows OS and App install disks and MS will quickly find that they no longer call a lot of the shots anywhere any longer .......

Versions of Chrome OS are available now that will run on re-purposed old MS machines, so hardware simply isn't the limiting factor that MS and Intel would wish for it to be.   This also says Wintel can't force you to go buy a new machine quite so easily any more, as Chrome OS is so light and fast it runs really really well on old generation Windows machines, just like Linux is willing to do.



===================================================



Now toss in the Linux idea of Snaps and Flatpacks and realize that devoted enthusiasts in the major distro worlds will be putting together flat packs and snaps to do ANY Windows program or game that your hardware can handle, no matter how stubborn that program might be to set up.  

You will be able to go get a snap or a flatpack from off the internet, make it legal by the existence of your software discs and just go DO IT.

Legalities can always get manipulated by MS and it is expected that all programmers who can do so will go to a freeware Windows world alternative such as REACTOS or LINUX/WINE for the Windows type operating portions that Chrome OS cannot supply.  

Certainly MS is capable of tweeking their OS literally overnight to stymie an offending Snap or Flatpack, but MS but cannot easily do so if it is really Reactos/Linux/Wine that is operating under the Snap/Flatpack's covers.
Back to top
 
« Last Edit: 08/07/18 at 19:46:58 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Oldfeller--FSO
Serious Thumper
ModSquad
*****
Offline

Hobby is now
"concentrated
neuropany"

Posts: 12637
Fayetteville, NC
Gender: male
Re: 2018 -- new Intel failures & issues
Reply #28 - 08/08/18 at 01:21:54
 

Deep breath time ---- the Wintel Monopoly still exists and as long as MS keeps promoting a technical monopoly in the USA by all its various tricks and methods, then Intel and Microsoft will forever remain with us.    

Yes, AMD is becoming a strong competitor to Intel, but that in fact plays along with the Wintel monopoly as AMD and Intel share the load of the "physical PC" end of things, using the same MS born constraints.

The Chinese are breaking into the hardware design side more and more, but they tend to operate along existing MS guidelines as they have no special Chinese ideology to promote in particular.    They are just there to make a buck,  endlessly seeking the cheapest and FASTEST implementation pathways, and that generally comes with zero concern shown for infrastructure or any standards of any kind.

China's Government has mandated the 100% use of Linux before, and that has only lasted for a month or two before the unruly Chinese business people reverted to the path of least effort and "fastest results".   This was "use what you got" and that was Windows, by and large.

Google has amassed all of Android and all of Linux and has put them all within the Chromebook purview,  but this does not translate into more than an 11-16% per year growth in Chromebook market share increase.

Even Linux itself is based on re-purposing old Wintel equipment, and although Linux use is spiking at the moment that is due to the retirement of a vast fleet of old XP, Win 7, Win 8 machines carrying a variety of old OS variants.

Microsoft is not limited by the constraints they place on everybody else, so Mickey has now provided an older version of Win 10 to the refurb houses as a freebee so they can wipe all the old machines of all old OS versions and put the freebe Win 10 on all of them as a consistent "wipe and replace" on every hard drive --- giving all the refurbs "a new lease on life'.    

The advantage to MS is simple, getting Win 10 put on the hard drive means they get the chance to extort money from the new owners as they roll into the subscription plan era.


Scary, ain't it?         Roll Eyes
Back to top
 
« Last Edit: 08/08/18 at 11:02:25 by Oldfeller--FSO »  

Former Savage Owner
  IP Logged
Eegore
Serious Thumper
*****
Offline

SuzukiSavage.com
Rocks!

Posts: 8005

Re: 2018 -- new Intel failures & issues
Reply #29 - 08/08/18 at 07:10:13
 

 So other than the cost what is a negative to leasing PC's and the software from a company that will do all the work of keeping it maintained for me?

 This sounds like a cell-phone plan to me.
Back to top
 
 
  IP Logged
Pages: 1 2 3 4 ... 31
Send Topic Print


« Home

 
« Home
SuzukiSavage.com
04/25/24 at 22:55:43



General CategoryThe Cafe › 2020 -- new Intel failures & successes


SuzukiSavage.com » Powered by YaBB 2.2!
YaBB © 2000-2007. All Rights Reserved.