Nvidia Ampere Discussion [2020-05-14]

Discussion in 'Architecture and Products' started by Man from Atlantis, May 14, 2020.

  1. Man from Atlantis

    Regular

    Joined:
    Jul 31, 2010
    Messages:
    747
    Likes Received:
    60
    [​IMG]
    A100 SXM2
    I found the die size around 806mm2

    Edit: EETimes went ahead and published the article before its due

    Nvidia Reinvents GPU, Blows Previous Generation Out of the Water
    http://archive.is/fiMX1
     
    #1 Man from Atlantis, May 14, 2020
    Last edited: May 14, 2020
    Lightman, tinokun, nnunn and 4 others like this.
  2. xpea

    Regular Newcomer

    Joined:
    Jun 4, 2013
    Messages:
    404
    Likes Received:
    430
    PS:
    For Jensen personal usage :runaway::runaway::runaway:
     
    #2 xpea, May 14, 2020
    Last edited: May 14, 2020
  3. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    8,862
    Likes Received:
    2,792
    Location:
    Finland
    "Nvidia claims the A100 has 20x the performance of the equivalent Volta device for both AI training (single precision, 32-bit floating point numbers) and AI inference (8-bit integer numbers)." ....if you use new Tensor Float 32 -precision not supported by Volta
     
    Lightman, tinokun, TheAlSpark and 5 others like this.
  4. techuse

    Newcomer

    Joined:
    Feb 19, 2013
    Messages:
    220
    Likes Received:
    121
    Ya i knew there was some type of typical Nvidia shenanigans at play when i read that 20x blurb.
     
  5. DegustatoR

    Veteran

    Joined:
    Mar 12, 2002
    Messages:
    1,479
    Likes Received:
    219
    Location:
    msk.ru/spb.ru
    And it'll be "just" 2.5x if you don't.
     
  6. CarstenS

    Veteran Subscriber

    Joined:
    May 31, 2002
    Messages:
    4,913
    Likes Received:
    2,232
    Location:
    Germany
  7. PSman1700

    Veteran Newcomer

    Joined:
    Mar 22, 2019
    Messages:
    2,327
    Likes Received:
    720
    That UE5 demo is probably even more impressive compared to what NV can tech demo on a 3080/nvme optane system.
     
  8. xpea

    Regular Newcomer

    Joined:
    Jun 4, 2013
    Messages:
    404
    Likes Received:
    430


    it's a change from previous Volta / Turing gen

    another source:
    https://www.marketwatch.com/story/n...is-coronavirus-2020-05-14?link=MW_latest_news
     
    #8 xpea, May 14, 2020
    Last edited: May 14, 2020
  9. ToTTenTranz

    Legend Veteran Subscriber

    Joined:
    Jul 7, 2008
    Messages:
    10,840
    Likes Received:
    5,407
    I sure hope this server GPU isn't the only thing they're going to show today, and we'll be able to get a glimpse of the new consumer lineup.
     
  10. DavidGraham

    Veteran

    Joined:
    Dec 22, 2009
    Messages:
    3,254
    Likes Received:
    3,463
    So now that Ampere is the overall arch for consumer and HPC, consumer chips will most likely cut down on Tensor units count, there is also a possibility of a Titan Ampere GPU as well, like Titan V.

    The bigAmpere HPC is definitely a 128SM GPU, we need to figure out the frequency and power consumption now so we can infer some info about the rest of the lineup.
     
  11. szatkus

    Newcomer

    Joined:
    Mar 17, 2020
    Messages:
    17
    Likes Received:
    18
    To reach 5PTFLOPS it needs a bit more than 128SM. Or really high clocks.
     
  12. pjbliverpool

    pjbliverpool B3D Scallywag
    Legend

    Joined:
    May 8, 2005
    Messages:
    7,719
    Likes Received:
    928
    Location:
    Guess...
    I haven't done the math but is that based on CUDA FLOPS or Tensor FLOPS? As the Tensor cores pump out a lot more.
     
  13. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    8,862
    Likes Received:
    2,792
    Location:
    Finland
    8 GPUs pushing 5 PFLOPS is around 625 TFLOPS per GPU, clearly they're talking about Tensor-FLOPS and not general FP32 FLOPS
     
    Lightman, PSman1700 and BRiT like this.
  14. DavidGraham

    Veteran

    Joined:
    Dec 22, 2009
    Messages:
    3,254
    Likes Received:
    3,463
    Lightman, tinokun, nnunn and 3 others like this.
  15. xpea

    Regular Newcomer

    Joined:
    Jun 4, 2013
    Messages:
    404
    Likes Received:
    430
    DGX A100 specs
    DGX A100 specs.png
    AMD EPIC 7742 !!!
     
    Lightman, hurleybird, nnunn and 3 others like this.
  16. pjbliverpool

    pjbliverpool B3D Scallywag
    Legend

    Joined:
    May 8, 2005
    Messages:
    7,719
    Likes Received:
    928
    Location:
    Guess...
    Yeah I was being very lazy on the maths. This sounds correctto me then as the 2080Ti is rated at 440 TFLOPS in INT4 and that's with 80SM's. So it seems to be running at a slightly slower clock than the 2080Ti all else being equal.
     
  17. DavidGraham

    Veteran

    Joined:
    Dec 22, 2009
    Messages:
    3,254
    Likes Received:
    3,463
    So the first iteration: Tesla A100 is a cut down one, from 128SM to 108SM.
     
    pharma likes this.
  18. xpea

    Regular Newcomer

    Joined:
    Jun 4, 2013
    Messages:
    404
    Likes Received:
    430
    2020-05-14 21_13_19-NVIDIA GTC 2020_6_1.png

    Big numbers here !!!
     
    Lightman and pharma like this.
  19. manux

    Veteran Regular

    Joined:
    Sep 7, 2002
    Messages:
    1,946
    Likes Received:
    815
    Location:
    Earth
    Omniverse ray tracing stuff looks great. Link should be to right time, if not it starts around 9:42. The actual explanation what omniverse is before the demo part

     
    Sxotty, Lightman, pharma and 2 others like this.
  20. DavidGraham

    Veteran

    Joined:
    Dec 22, 2009
    Messages:
    3,254
    Likes Received:
    3,463
    So NVIDIA effectively traded FP32 CUDA cores with FP32 Tensor Cores, Tesla A100 is really just Ampere optimized for AI.

    Regular FP32 is : 19.5 TF
    Tensor FP32: 156 TF, accelerated to 312 TF effective through "sparse acceleration"

    Consumer Ampere will definitely cut down on the advanced tensor stuff and trade back the lost FP32 CUDA cores.
     
    #20 DavidGraham, May 14, 2020
    Last edited: May 14, 2020
Loading...

Share This Page

  • About Us

    Beyond3D has been around for over a decade and prides itself on being the best place on the web for in-depth, technically-driven discussion and analysis of 3D graphics hardware. If you love pixels and transistors, you've come to the right place!

    Beyond3D is proudly published by GPU Tools Ltd.
Loading...