Nvidia's Tesla P100 does not use the full Pascal GPU core
Nvidia's Tesla P100 does not use the full Pascal GPU core
Published: 7th April 2016 | Source: OC3D |
Nvidia's Tesla P100 does not use the full Pascal GPU core
Nvidia's Tesla P100 does not use the full Pascal GPU core, with 256 GPU cores being left out of the design. This many mean that "Big Pascal" may be able to offer a full 11.3TFLOPs of single precision performance.
Right now Nvidia has been very clear that the Pascal-based Tesla P100 uses only 56 SM units, but when looking at Nvidia's pascal GP100 core design we can see that there is a total of 60 SMs available, meaning that there are a full 256 Pascal GPU cores that will be unavailable in the Nvidia P100.
Nvidia may not be using the full GP100 pascal GPU core in the Tesla P100 for many reasons, but it is very likely that it is due to yield issues with the new GPU design. The Nvidia GP100 core is the largest that I have ever seen from Nvidia, with a die size that is 610mm squared, making it larger than even the Titan X's 601mm squared core design. This GPU is also the first Nvidia GPU that has been shown that was made using the TSMC 16nm FinFET processing node and the first to use HBM 2.0 memory, making this new GPU core design much more complex to manufacture.
This GPU is also the first Nvidia GPU that has been shown that was made using the TSMC 16nm FinFET processing node, which is a new processing node from TSMC, and is the first Nvidia GPU to use HBM 2.0 memory, making this new GPU core design much more complex to manufacture.
With HBM 2 being a larger size than HBM 1 per chip and Nvidia's GP100 GPU core being so large it looks like it will be a very difficult task to fit all of this hardware on a single interposer, likely stretching the current tolerances of this technology to their limits. Just look at the picture below to see how close the four HBM 2.0 chips are to the GPU core if you want to see how tightly packed this GPU design is.
Below we have looked at the specifications of Nvidia's Tesla P100 and Nvidia's GP100 core design to list the specifications of Nvidia's full GP100 GPU core though many of the specifications remain unknown at this point.
Right now we do not know if Nvidia will use larger capacity HBM 2 chips on GP100 in the future, as we do know that 8GB HBM chips are due to be manufactured this summer allowing a potential 32GB GP100 based GPU to be released in the future.
We also cannot comment on what clock speeds will be possible in future GPUs using the GP100 GPU core, as we do not know if Nvidia's stated 1328MHz/1480MHz base and boost clocks are conservative on Nvidia's part or not. For all we know later Pascal chips could run at 1600MHz+ boost in the future if the chip overclocks well.
Tesla K40 | Tesla M40 | Tesla P100 | Full GP100 | |
GPU Core | GK110 (Kepler) | GM200 (Maxwell) | GP100 (Pascal) | GP100 (Pascal) |
SM Units | 15 | 24 | 56 | 60 |
Cores Per SM Unit | 192 | 128 | 64 | 64 |
FP32 CUDA Cores | 2880 | 3072 | 3584 | 3840 |
Base Clock Speeds | 745MHz | 948MHz | 1328MHz | - |
Boost Clock Speeds | 810/875Mhz | 1114MHz | 1480MHz | - |
Texture Units | 240 | 192 | 224 | 240 |
VRAM Cappacity | Up to 12GB | Up to 24GB | 16GB | - |
Memory Interface | 384-bit | 384-bit | 4096-bit | 4096-bit |
Transistors | 7.1 billion | 8 billion | 15.3 billion | 15.3 billion |
Die Size | 551mm^2 | 601mm^2 | 610mm^2 | 610mm^2 |
Manufacturing Process | 28nm | 28nm | 16nm | 16nm |
Right now Nvidia states that the P100 is in volume production and that the chip will be selling first to the supercomputing market in June 2016 and will be arriving with OEMs in Q1 2017.
Nvidia has stated that they are not currently planning on releasing the GP100 GPU core in a Geforce GTX part but given the fact that the GK110 and GM200 GPU cores eventually became GTX Titan series GPUs and later GTX X80 Ti series GPUs it seems likely that this GPU will eventually hit the market as a future flagship GPU when production yields and volume make it affordable.
You can join the discussion on Nvidia's full GP100 GPU core specifications on the OC3D Forums.
Nvidia's Tesla P100 does not use the full Pascal GPU core. Full core details inside.https://t.co/tltLnGajqA pic.twitter.com/g0I6fKvxM0
— OC3D (@OC3D) April 7, 2016
Most Recent Comments
There could be a full fat version later with everything enabled.
HBM2 and NVlink will probably stay as it would mean redesigning the chips to get rid of them.
Titan card will probably have all the cores enabled and the Ti version will possibly have specs similar to the OP but with higher clockspeeds.
Don't think gamers will see any of this until 2017.Quote
Probably down to yields on early production.
There could be a full fat version later with everything enabled. HBM2 and NVlink will probably stay as it would mean redesigning the chips to get rid of them. Titan card will probably have all the cores enabled and the Ti version will possibly have specs similar to the OP but with higher clockspeeds. Don't think gamers will see any of this until 2017. |
I think this time they will make a bigger gap between Titan and Ti. They should at least, people will pay for it and make a bigger profit so business wise it would work.Quote