Wednesday, May 10th 2017
NVIDIA Announces Its Volta-based Tesla V100
Today at its GTC keynote, NVIDIA CEO Jensen Huang took the wraps on some of the features on their upcoming V100 accelerator, the Volta-based accelerator for the professional market that will likely pave the way to the company's next-generation 2000 series GeForce graphics cards. If NVIDIA goes on with its product carvings and naming scheme for the next-generation Volta architecture, we can expect to see this processor on the company's next-generation GTX 2080 Ti. Running the nitty-gritty details (like the new Tensor processing approach) on this piece would be impossible, but there are some things we know already from this presentation.
This chip is a beast of a processor: it packs 21 billion transistors (up from 15,3 billion found on the P100); it's built on TSMC's 12 nm FF process (evolving from Pascal's 16 nm FF); and measures a staggering 815 mm² (from the P100's 610 mm².) This is such a considerable leap in die-area that we can only speculate on how yields will be for this monstrous chip, especially considering the novelty of the 12 nm process that it's going to leverage. But now, the most interesting details from a gaming perspective are the 5,120 CUDA cores powering the V100 out of a total possible 5,376 in the whole chip design, which NVIDIA will likely leave for their Titan Xv. These are divided in 84 Volta Streaming Multiprocessor Units with each carrying 64 CUDA cores (84 x 64 = 5,376, from which NVIDIA is cutting 4 Volta Streaming Multiprocessor Units for yields, most likely, which accounts for the announced 5,120.) Even in this cut-down configuration, we're looking at a staggering 42% higher pure CUDA core-count than the P100's. The new V100 will offer up to 15 FP 32 TFLOPS, and will still leverage a 16 GB HBM2 implementation delivering up to 900 GB/s bandwidth (up from the P100's 721 GB/s). No details on clock speed or TDP as of yet, but we already have enough details to enable a lengthy discussion... Wouldn't you agree?
This chip is a beast of a processor: it packs 21 billion transistors (up from 15,3 billion found on the P100); it's built on TSMC's 12 nm FF process (evolving from Pascal's 16 nm FF); and measures a staggering 815 mm² (from the P100's 610 mm².) This is such a considerable leap in die-area that we can only speculate on how yields will be for this monstrous chip, especially considering the novelty of the 12 nm process that it's going to leverage. But now, the most interesting details from a gaming perspective are the 5,120 CUDA cores powering the V100 out of a total possible 5,376 in the whole chip design, which NVIDIA will likely leave for their Titan Xv. These are divided in 84 Volta Streaming Multiprocessor Units with each carrying 64 CUDA cores (84 x 64 = 5,376, from which NVIDIA is cutting 4 Volta Streaming Multiprocessor Units for yields, most likely, which accounts for the announced 5,120.) Even in this cut-down configuration, we're looking at a staggering 42% higher pure CUDA core-count than the P100's. The new V100 will offer up to 15 FP 32 TFLOPS, and will still leverage a 16 GB HBM2 implementation delivering up to 900 GB/s bandwidth (up from the P100's 721 GB/s). No details on clock speed or TDP as of yet, but we already have enough details to enable a lengthy discussion... Wouldn't you agree?
103 Comments on NVIDIA Announces Its Volta-based Tesla V100
Somehow i feel like Nvidia has already maxed out the possible efficiency optimization for Pascal-Maxwell CUDA designing. They are also back to the MOAR CORE and Higher MHz direction. With so many CUDA units available i am pretty sure Async Computing will be Volta's advantage. It should perform pretty well in Vulkan and DX12.
Poor VEGA.
Is it me or have they done little but rearrange the graphics components then add a shit tone of ai processors and call it done.
I am now more interested in lower tier volta because this ones for egg heads :) and not so much gamers ,what the heck is small volta going to look like given big voltas trying to go all Cyberdyne on us.
I don't believe this is going to be easy to market to joe public though because any record for dearest consumer card is going to get obliterated when this hits the shops Christmas 2018 ;).
devblogs.nvidia.com/parallelforall/inside-volta/
Ouch.
Nvidia has give a glimpse on Volta, meanwhile AMD still teasing us with Vega :shadedshu:
You never saw the GP100 for consumers and you will never see this. Expect a completely different card with no FP16, FP64 and Tensor-capabilities.
It also has the capability of executing INT32 and FP32 simultaneously. I know devs at my work are already frothing in their mouths just reading about it.
And I'm not sure where you got the 6% from.
Strip away the tech not relevant to gaming and GV104 and GV102 chips are gonna absolutely fly.
It's not like Pascal has any competition yet anyway. Can i say Happy Birthday GTX 1080 yet?
But the simple truth is users are users and when competing for them, it's the companies that are expected to bend over backwards. Crying like a baby that your good product doesn't sell is not a market strategy.
Would be nice if the 2070 would come close to 1080TI levels ...
Edit: Cannot wait for Vega, though. Really wanna see what AMD has to offer as well.