Intel's Knights Corner Turns Into The Xeon Phi

[droidhacker]

To all of the people spreading FUD that this card is less powerful than a 5 year old desktop ATI part... GET THE FACTS. People here are comparing theoretical hand-wavy peak performance numbers for *single precision operations* that are never reached in real life to actual certified *real* performance numbers for actual benchmarks achieved in the MIC systems. There is a *WORLD* of difference and if you don't believe me, go read the TOP 500 list and see how many 5 year old AMD GPUs are in there if those parts are supposedly amazing... I can save you some time since the answer is 0.

Let me put it to you this way: The exact same benchmarks where Intel has already shown MIC working at over 1 Terraflop are the benchmarks where Nvidia is *projecting* that its *full GK110 part* will be when it is finally released. Basically, Nvidia's top-of-the-line next-generation 7+ Billion transistor monster will be in the same league as MIC, but require you to use the CUDA programming model to get the performance. MIC totally destroys any existing compute accelerator on the market, and as a huge bonus the programming model for MIC is light years ahead of having to use CUDA or whatever passes for OpenCL these days in AMD land. The MIC is a fully documented architecture that supplies SIMD instructions that are expanded from the existing AVX instructions already used in Intel & AMD CPUs.

MIC is a *vastly* more open architecture than anything from Nvidia or AMD. And don't even get me started with the likes of Quaridiot who act like the Messiah has returned when AMD releases incomplete and inaccurate docs for some of its cards long after they have been released for a couple of unpaid volunteers to decipher. MIC is a 100% open documented architecture and Intel has already released open source software for it in advance of its launch. This architecture will hopefully force Nvidia and AMD to *really* take Linux seriously instead of treating it like a second class citizen while trying to make huge $$$ on Linux-based HPC systems.

So if the "facts" come from intel, they're truth, if they come from anyone else, they're lies? Get your bullshit straight. Intel is competent at precisely one thing; faking benchmarks.

So if you want to get your FACTS, buy one and run some *REAL* code on it. See it fall on its face.

[popper]

This is exactly what I was thinking several years ago.
Central CPU will be managing tasks, where pluggable CPU modules will provide actuall performance.
Multiply these modules with OpenCL and you have ray-trace real-time graphics.

don't forget your history that already happened Many many years ago originally in the BBC and AMIGA add in many core Transporter boards, its taken all this time to become popular again, but it may finally go somewhere on mass with Intel behind this new OLD UK Transporter concept thats come to the x86 landscape now....

[popper]

From what I've heard, the Larabee/KC/Phi design should scale better with branchy code, which is something that GPUs really suck at.

For certain problem sets GPUs are really good and you can get something approximating its maximum stated performance, but the moment you start adding branches that start sending threads in different directions your performance takes a nosedive.

Veerappan, will the chemical company you now work for be buying some of these knights corner ( they should have kept that name its more memorable than phi) for you to play with and try ?, perhaps you should ask them to then you can have a go at writing some 512bit AVC? SIMD patches for your beta OpenCL code base.... and post results....

and if they do, it might be nice to see you also wrote a few assembly and C patches for x264 to see the checkasm results of a large 1080p upscale to 4k/8k rescale and encode on these too if they get you some boards, and/or put one online and give the x264 dev's remote access to write and run hd and super HD tests for it.

assuming their new many core 512 SIMD is up to scratch for generic use and not just some oddball uni niche and top 500 companies with money to burn for overpriced industrial Co-Processor cards.

[popper]

don't forget your history that already happened Many many years ago originally in the BBC and AMIGA add in many core Transporter boards, its taken all this time to become popular again, but it may finally go somewhere on mass with Intel behind this new OLD UK Transporter concept thats come to the x86 landscape now....

that OC should read Transputer

http://en.wikipedia.org/wiki/Transputer

http://www.classiccmp.org/transputer/

[Kano]

I dont know if amd developed ecc ram support or not, if not the chips will not be used for super computing. The tflop speed of that new intel card seems to very impressive, nvidia will have to work hard to make their new gk110 chip faster. That card is most likely useless for desktop users but for number crunching it could be very successful. Also intel can usually produce a huge number of chips without problems as they do not outsource manufacturing. When tsmc has got issues nvidia (and amd) can not sell so many chips they want. The gk104 was needed to optimize the production process at tsmc so that gk110 can be build later. So in theory intel has got good arguments when they want sell the new coprocessor cards.

[Veerappan]

Veerappan, will the chemical company you now work for be buying some of these knights corner ( they should have kept that name its more memorable than phi) for you to play with and try ?, perhaps you should ask them to then you can have a go at writing some 512bit AVC? SIMD patches for your beta OpenCL code base.... and post results....

and if they do, it might be nice to see you also wrote a few assembly and C patches for x264 to see the checkasm results of a large 1080p upscale to 4k/8k rescale and encode on these too if they get you some boards, and/or put one online and give the x264 dev's remote access to write and run hd and super HD tests for it.

assuming their new many core 512 SIMD is up to scratch for generic use and not just some oddball uni niche and top 500 companies with money to burn for overpriced industrial Co-Processor cards.

Wait... I work for a chemical company now? Would've been nice if someone told me so that I wouldn't be showing up every day somewhere else.

I don't think I'll have any of these cards to play with any time soon, as I'd have to purchase it using my own money. For now, I'll just keep waiting on CL support for r600g to stabilize (and pitch in where ever I can). If the Knight's Corner/Phi stuff takes off, I'll hope the boards get cheap enough for us mortals to purchase. And well, I honestly don't care much about x264. I like what they're doing, but I'm not going to go out and specifically write code for a patent-encumbered video codec with an uncertain future (e.g. player royalties could start becoming a reality in the next few years).

[popper]

Wait... I work for a chemical company now? Would've been nice if someone told me so that I wouldn't be showing up every day somewhere else.

I don't think I'll have any of these cards to play with any time soon, as I'd have to purchase it using my own money. For now, I'll just keep waiting on CL support for r600g to stabilize (and pitch in where ever I can). If the Knight's Corner/Phi stuff takes off, I'll hope the boards get cheap enough for us mortals to purchase. And well, I honestly don't care much about x264. I like what they're doing, but I'm not going to go out and specifically write code for a patent-encumbered video codec with an uncertain future (e.g. player royalties could start becoming a reality in the next few years).

oh i just took your "I've settled into a position with a pharmaceutical research software company. I'll still keep poking around at the VP8/CL code as time allows" as working for a large chemical/pharmaceutical research software company that go through these kinds of Co-processors in bulk for daily use in their molecular modeling etc... oh well , its different if your paying for these things yourself, fair enough.

no problem there will be lots of dev's buying and doing that patching/testing anyway if this 512bit SIMD AVX (not avc OC ) is any good as Ultra High Definition is officially on the landscape now, i just thought i would ask just in case anyway.

[Wyatt]

It's an interesting architeture, and I know at least a couple people who are pretty stoked about it. I'm not sure how I feel about x86 living on even still, but Intel somehow has made it work.

[ldesnogu]

This chip won't run all x86 code: no 32-bit mode, no MMX/SSE/SSE2/etc (replaced with wider SIMD units), no CMOV. In particular code that uses intrinsics or assembly routines making use of SSEx will have to be rewritten.

All in all being x86 here doesn't add anything except that it helped Intel porting Linux and their C compiler.

Also people should stop dreaming: being able to efficiently use more than 50 cores is a wet dream given how most existing software can't even efficiently use our 4 core CPU.

So this is an interesting chip, but for HPC and rendering, not for much else.