OF, I know you don't like win 10 , but I tried this and got a free upgrade from win 7. It hung up for a day, but I turned off update and the process completed. Just thought some people might find this useful

https://www.forbes.com/sites/gordonkelly/2020/02/04/how-to-upgrade-to-windows...

Intel cannot downshift their lithography size so they are cutting their prices some more yet again.   Intel is adding in some additional "somewhat bogus" hyper-thread count talk-talk so they try to can claim some sort of equivalency with AMD's current chipsets (only by using special Intel test suites -- i.e. pure Intel BS testing on experimental chips that are not actually real or available).

AMD still has a real physical cost and speed and thread count and data throughput advantage at each comparative level of chipset.
You still have to double your Intel cost to get close to the equivalent REAL AMD functionality.

Hard fact, AMD is simply better stuff now when compared to Intel "apples to apples"  .....

Intel's current PR trick is to take the specifications off of item Intel B, the calculation ability off of Intel item C and then actually go sell you an older version chipset, a lesser lithography level chipset D ---- all the while headlining and only talking about a proposed brand new Intel chipset A, one that you cannot buy at all until late next summer.

To say Intel is brewing up some intentional PR confusion is only the simple stark truth now-a-days.

This illegal and forbidden practice is called "bait and switch" when it takes place in the automotive industry.   Fines and jail time are assessed for the folks doing it to automotive consumers because laws were passed on the subject long ago.

Although Intel talks about 4 hyper threads per core, Intel cannot really actually do that trick yet.   Only certain AMD processors running on Linux can actually do that trick for real right now and the workloads that can actually use it live on rackspace units, not on any PC units.

Clear Linux will get this capability first for any of the Intel chipsets, maybe by late this summer.   MS Win 10 hasn't moved off dead center about adopting 4 threads per core yet at all, and frankly may never do so as MS makes separate OS products for rack units and MS does not intermix features with the PC based units.



All Intel is really going to accomplish with all this blather is to motivate AMD to change over to 5nm lithography even quicker ........

On the positive side, AMD inside of two years has taken Intel up from 4 cores across the board up to 6-8 cores now, with 10 cores planned by Intel "for late next summer" .......  and forced Intel to lower their prices by 25% to get down to a not quite competitive posture.

AMD has already totally fixed their "hyperthreading" so it is not an AMD security concern at this point in time, Intel has not yet completed their first wave of hyperthreading fixes and Intel hyperthreading still is very much considered a security concern.

https://www.extremetech.com/computing/306978-intel-expects-to-reach-process-p...



According to remarks Intel CFO George Davis made at a Morgan Stanley conference this week, the company still believes it has a ways to go before it matches the pace of its foundry competitors and retakes overall process leadership. Reports of his remarks at the conference suggest Intel won’t regain parity with TSMC and possibly Samsung until it launches 7nm parts in 2021, with the firm retaking leadership at the 5nm node. Intel has previously said it would launch a GPU on 7nm in 2021, and Intel CEO Bob Swann has stated that 7nm CPUs will ship in Q4 2021

OK, Intel has been letting their acting Chief Financial Officers speak for the company as "current spokesman" again.   This new one is called George Davis, the new Corporate CFO.

This is two bean pickers in a row promoted to be "acting talking head" for the company  with Bob Swann now moving back up a step into the exalted Chairman of the Board CEO status and the acting current Intel Finance type bean picker taking up the main public "talking head" position.



Reality vs Intel    This year is 2020, first quarter.


The world is ACTUALLY AT 7nm right now and is busy stepping down to 6-5nm as we speak.  

6-5nm will roll across the board everywhere this year, everywhere but at Intel and Global Foundry.

In 2021, second gen 5nm will hit, and possibly 3rd gen 5nm if it rolls out as fast as the first ones did.

2022 is likely the start of 3nm turf (assuming the two years per lithography generation as driven by Apple holds true).

Intel's head dog info from both head bean pickers says that Intel WILL REMAIN at least two generations back and as such Intel is being counted out as relatively mortibund by doing so.



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



https://www.anandtech.com/show/15253/80core-n1-nextgen-ampere-quicksilver-the...



The new product doesn’t yet have a marketing name – we asked and they preferred it to be called ‘Next-Gen Ampere’ for now, or to use its SoC codename ‘QuickSilver’. What we were told is that the new product is a brand new ground-up design from Ampere, separate from the AppliedMicro IP acquisition. It plans to compete in the same space that Amazon’s Graviton2 currently sits at AWS, but as the main alternative to the other cloud providers that won’t have access to Graviton2.

What we are getting today is some rough details of the new chip. Other features, such as exact SKUs to be launched, exact TDPs, exact frequencies, and pricing, are going to be disclosed at the official release announcement in 2020. Nonetheless, Ampere has exposed a lot of details.

Next-Gen Ampere will be a monolithic chip built on TSMC’s 7nm process and featuring 80 cores. These cores are not custom like eMAG, but are built on Arm’s Neoverse N1 design, using paired clusters of cores connected by an Arm mesh IP (CMN-600). This is the same core as Graviton2, and as expected Ampere is keen to promote that their design is optimized for power, performance, latency, and throughput, as well as offering more cores and more of other things as well.

ARM Holdings has put out a 7nm factory standard N1 design version of the ARM cores with a heavy duty infinity fabric data buss and AI accelerators that is primarily intended for use in Data Centers.   Companies have put together 36 core, 64 core and now 80 core versions of this product and have begun sampling and testing.  Expect this tech to roll downhill to the PC world within several years or so .....

This 80 core N1 variant has now been sampled and is out at several of the data farm folks being tested for its claims of 2x+ processing power at <1/2 the energy cost of the best of the Intel mainframe chipsets.   The cost of the processor is low enough to get down into upper range PC space right now, and if it becomes common place the price will drop further.

Remember, the biggest data crunchers -- Google, Amazon, etc. etc.  already roll their own arm based processors and are currently reaping these sorts of cost and speed benefits from doing so.  

So, we already know the idea works.


Prediction time

AMD has taught the world (and the OS boys) that lots of cores are cool and are a perfectly acceptable way to get the job done faster and at the lowest cost.

I have an 8 core Motorola cell phone swinging 8 tiny energy efficient A53 cores that uses this "many little processors" method to get the job done, and that includes doing video playback tasks that used to only run off a PC box.

Trust me on this --- 24 or 36 or 64 or 80 ARM Holdings N1 cores can get the same job done on a PC replacement chipset.

If AMD drops their current quite slow "sandbagged" improvement pace any further due to Intel not moving any at all, then the ARM processor boys will pick up their x86 fumble AT ONCE and run the dropped ball on down the field in the opposite direction using the ARM Holdings N1 cores and a consumer version of the new mainframe connection buss fabric.

Intel is no longer relevant and is quite moribund and immovable -- AMD needs to watch out for their real current "phone based" competitors sitting over there against the wall jest a glaring at you and fondling their N1 licenses --- and yes, they are glaring at you AMD you good old buddy you .......

https://www.extremetech.com/computing/307097-amd-analyst-day-2020-zen-3-infin...

HOW GOOD ARE THE NEW RYZEN 4000 ALL IN ONE CHIPSETS?



Note please that this is compared to an Intel i7 chipset that you can't really buy just yet, certainly not in just any run of the mill Intel laptop to say the least.

This is right now sort of stuff that AMD is shipping right now.

Intel promises to do better by next summer ........ but by the time it arrives AMD will have rolled out the entire Zen 3 generation for laptops and will have lapped whatever Intel does do very strongly once again.

Intel's best guess at a cadence is a tick tock two year cycle right now, assuming they can actually make their changes work right on a two year schedule.    So far, Intel hasn't been able to do this in house and has been buying their 7nm laptop tech from Samsung and TSMC on the sly.

AMD is doing a tick tick tick cycle inside of one year right now, with whole new lithography generations coming out of TSMC within a single year.   AMD is hitting the current tick with the next one of their product groups, skipping between the groups on a rolling basis and simply murdering Intel on product advancements.

Watch AMD begin consistently skipping the lead in first efforts of the TSMC cycles and just hitting on the first refinement cycle rather than fighting out the details of those first "somewhat bug ridden" first of the lot lithography generations.

Watch the ARM phone based boys become the defacto state of the art technical leaders using all the new ARM N1 processor generations and all the various new AI stuff --- becoming the major PC suppliers to China and to India while they do that.

https://optocrypto.com/intel-will-use-6nm-tsmc-nodes-in-2021-and-3nm-in-2022/



As for semiconductor technology, Intel’s 10 nm was mass-produced just once with very mixed results, but the company also has stated that its internal production yields and capacity will not be as large as Intel 22 nm and Intel 14 nm, which could be an important sign. That’s why Intel is now seriously thinking about TSMC and will deploy its 6 and 3 nm nodes in the coming years.

Intel would outsource chips to TSMC with 6 and 3 nm nodes
Previously, the industry had repeatedly reported that Intel would also outsource chips to TSMC. According to the latest information, this will extend to 3 nm in 2022, after the 6 nm node in 2021.

Intel expects to fully exploit the 6-nanometer TSMC process in 2021 and is currently testing it.

If the company really intends to expand the outsourcing of its chips, the GPU should be used first, in addition to the partially outsourced chipset, because the GPU is easier to manufacture than a CPU, and TSMC has experience in manufacturing GPUs.

Intel will use 6nm TSMC nodes in 2021 and 3nm in 2022
Intel’s Xe architecture alone shows that DG1 is manufactured in a proprietary 10 nm process. It has 96 execution units with a total of 768 cores, a basic frequency of 1 GHz, an acceleration frequency of 1.5 GHz and 1 MB cache as well as 3 GB video memory.

It is expected that the DG1’s performance will be comparable to that of a GTX 950, which is about 15% worse than the GTX 1050, a low-end graphics card suitable for energy-efficient areas, especially notebook GPUs.

After the DG1 comes the DG2. It has been previously reported that DG2 will use TSMC’s 7nm method. It is now possible that it will end up using the 6 nm.

The popular semiconductor manufacturer had also announced that Ponte Vecchio graphics cards for data centers will use their own 7-nm EUV process, we do not know if this plan will remain the same or will be switched to a 6-nm node. We will keep you informed.


OK, this has been rumored repeatedly for the last year, with some confirmations that Intel is indeed showing things at shows that were built by Samsung and some things being shown are now being built up by TSMC, all of which have a strong ARM knowledge underpinning and are not x86 based at all.

This is also backed up by Intel letting go a bunch of their x86 engineering staff (all experts in x86 tech but pretty much clueless about doing ARM designs).

Look to see Intel glomm on to the ARM N1 wave as their next technological "forward push".   This allows Intel to shake all the security holes that they cannot figure how to patch and to offload their "lack of expertise" on to the shoulders of their Asian suppliers just like Apple does.


"If you can't beat them, join them ......"




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


First confirmation from mainline sources (WCCFTECH and NOTEBOOKCHECK) saying Intel is moving on past their 10nm and 7nm in-house efforts, mainly because Intel's in-house yields are not good and the in-house processor performance isn't up to par either.

https://wccftech.com/intel-abandoning-10nm-after-dg1-planning-to-use-tsmcs-6n...

https://www.notebookcheck.net/Intel-s-upcoming-6nm-and-3nm-Xe-GPUs-to-be-buil...

Look to see Intel to get in line with the ARM based boys, adapting their TSMC ARM tech and using Clear Linux functions more and more as the basis for Intel's competitive future plans ............


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


INTEL CHEATS AGAIN

Just like they paid CTS to bushwack AMD with zero warning, Passmark has been contracted by Intel to give a paid victory to Intel over the new Passmark version 10 single thread performance testing.

In the last half year, AMD had wrested all the performance crowns from Intel the hard way, by outperforming Intel over a variety of measurement methods.

Intel found this was cutting into sales and that they were actually losing ground in gaming (games are all single tread work), so Intel reached out to Passmark to redo the Passmark test methodology to use an INTEL ONLY tool set when doing the test.

Version 9 of Passmark was in line with all other tests and showed AMD to be clearly ahead of Intel on single and multithreaded applications.   Version 10 of Passmark out just last week shows ONLY INTEL PROCESSORS in the top 50 places, as only Intel supports the tool set Passmark is using in Passmark version 10.

Gamers are beating up on Passmark for being stupid lazy shills, which both Passmark and CTS both certainly deserve for simply being stupid stupid stupid stupid lazy shills .......

Passmark's head dog apologies publicly for the Intel tool exclusivity, he didn't recognize that non-Intel tools were "as good as what Intel provided".   Says that version 11 of Passmark will attempt to be tool set agnostic and will simply track who completes the task sets first (like Passmark 9 did).

Passmark had taken the track with Passmark 10 that the best "state of the art" Intel tools needed to be used as Intel's "proper performance" required those tools.

Once again, Intel can't compete, so they simply pay folks to CHEAT for them ......

With all the Passmark controversy still belling in waves of confusion, and with Intel's massively misleading advertising and processor name swapping still happening (ongoing) and now the use of Intel proprietary (unreleased, non-industry standard) drivers --- WHAT IS WHAT IN PROCESSORS ????

This question has been addressed by several of the major testing reviewers as an item that was worth their time to answer.

First of all, unless you are buying a new computer this week --- forgetaboutit.  It will all change in the next 60-90 days on several fronts.

Right now AMD is inching ahead of Intel in real terms but the gap isn't stable as Intel is playing lots of different games to either be (or seem to be) equivalent in real performance.   Intel keeps "announcing a new product of the day" that isn't ever built or it gets replaced by another new product of the day very quickly.   All of these are 14nm products and none of them are both efficient AND fast at the same time.  

With Intel it is either/or, always.   With AMD, you get both all the time.

If you can use multiple core performance -- AMD is ahead of Intel and is going to stay there.   AMD is marching down the TSMC lithography drops as they happen, Intel is still stuck on their own 14nm processes.

If you only use single core performance (older gaming uses this) your older Intel machine is still fine, keep the old machine and use it until it dies.


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


What happens next?    AMD goes to 7nm Gen 3 this year and 5nm Gen 1 next year.   5nm performance increases will be major major performance increases coming in with significant cost reductions for the raw chip production.   Intel has nothing significant to say in response to this in any roll out plans that they have put out for the next 2-4 years.  

Given how weak Intel's execution has been in the last 4 full years, this is simply bad news for Intel.   Intel talks a lot about buying chipsets from TSMC "in the future" but in a company run by bean pickers this likely will never happen as Intel has too much profit invested in 14nm manufacturing to walk away from it (especially if led by a pair of bean pickers).

Intel has now opened the Huawei Pandora's box and pulled out the "progress by specialty drivers" trick.   Intel will play this new trick for all it is worth, but please understand that the Linux world actually maintains a set of industry standard drivers for everybody's equipment within the Linux kernel itself, these are the true industry standard drivers that generally can duplicate any "progressive tricks" anyone can come up with within a year or so and in plain reality simply makes them part of the real international industry standard.

Apple is joining the fray within 1-2 years with their A-14 series.   ARM HOLDINGS is joining the fray now with their mainframe originated 36-96 core A12 based multi-core N1 processors.   China is moving forward as a chip supplier to the world using older AMD x86 tech and modern ARM Holdings tech as their x86 and ARM basis.   We have waves of progress coming that will march down the lithography curve as it really happens.

Intel is currently static, Intel is not moving forward significantly right now at all.    Look for ALL of the current set of competitors to lap Intel inside the next 2 calendar years despite the Intel trickery.

OF, I'm using an HP p6320a, which I brought April 2010, it has an Intel ES5400 chip which was released about 1st quarter 2009. I'm now running Linux Mint Mate Sonya almost exclusively, for about a year haven't yet got rid of WIN7 but may do so in a few months, once I get someone to put the photos on Win 7 into Linux, have a brand new HP laser printer which prints well on Linux and scans ok?? I dont game and don't download that much, no one has hacked my bank account, so should I be that worried about an Intel chip?