Nvidia Ampere Discussion [2020-05-14]

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

  1. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,070
    Likes Received:
    2,942
    Location:
    Finland
    It varies from card to card a lot, even 100s of MHzs depending on cooling and other factors. And of course, different loads get different boosts
     
  2. DavidGraham

    Veteran

    Joined:
    Dec 22, 2009
    Messages:
    3,278
    Likes Received:
    3,523
    My MSI Ventus 2080Ti (maybe the cheapest 2080Ti ever) boosts to 1775MHz all the time.
     
    PSman1700 likes this.
  3. pharma

    Veteran Regular

    Joined:
    Mar 29, 2004
    Messages:
    3,528
    Likes Received:
    2,215
    Newguy, jlippo, Lightman and 4 others like this.
  4. PSman1700

    Veteran Newcomer

    Joined:
    Mar 22, 2019
    Messages:
    2,498
    Likes Received:
    775
    NV's TF numbers are very conservative for that reason. Almost 1800mhz on a high-end chip like that, damn.
     
  5. Man from Atlantis

    Regular

    Joined:
    Jul 31, 2010
    Messages:
    753
    Likes Received:
    79
    [​IMG]

    A100 PCIE variant, 250W TDP for %90 performance.
    https://www.nvidia.com/en-us/data-center/a100/

    https://www.computerbase.de/2020-06/nvidia-ampere-a100-pcie-karte-server/


    [​IMG]
    https://www.servethehome.com/new-nvidia-a100-pcie-add-in-card-launched/
     
    #305 Man from Atlantis, Jun 22, 2020
    Last edited: Jun 22, 2020
    Lightman, TheAlSpark, BRiT and 3 others like this.
  6. DavidGraham

    Veteran

    Joined:
    Dec 22, 2009
    Messages:
    3,278
    Likes Received:
    3,523
    So as explained before, the SXM4 form factor extends power to 400W to accommodate NVLink and higher sustainable clocks, V100 had the same thing happen to it when it extended the SXM form factor to 350W and one more time for 450W.
     
    pharma, Benetanegia and nnunn like this.
  7. jayco

    Veteran Regular

    Joined:
    Nov 18, 2006
    Messages:
    1,264
    Likes Received:
    562
    https://hardwareleaks.com/2020/06/21/exclusive-first-look-at-nvidias-ampere-gaming-performance/

    3080 +30% over 2080TI?

    Nvidia doesn't give TF numbers for their consumer cards. For the same reason reviewers don't rely on TF numbers to do their comparisons. Or for the same reason a Radeon VII (14TF) is barely able (in some games) of beating the 2080 (10TF).
     
  8. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,070
    Likes Received:
    2,942
    Location:
    Finland
    Last page called, it wants the link back

    Don't be ridiculous, they stopped reporting TFLOPS only when they launched RTX-generation because the new fancy "RTXOps" they invented makes the difference look far bigger than TFLOPS would.
     
    Lightman, entity279, techuse and 2 others like this.
  9. neckthrough

    Newcomer

    Joined:
    Mar 28, 2019
    Messages:
    14
    Likes Received:
    31
    It's fair to critique the precision and utility of the specific "RTXOps" metric. However we can't avoid the uncomfortable reality that the utility of raw TFLOPs as a single metric to judge performance is diminishing as chips move towards more specialization in a post-Moore's Law world. Whether it's even possible to come up with a vendor-agnostic and cross-benchmark metric is a tough question. Even "FPS in game XXX" is becoming more challenging due to the difficulty in performing iso-quality comparisons because of divergent feature sets.
     
    Geeforcer, Rootax, xpea and 3 others like this.
  10. jayco

    Veteran Regular

    Joined:
    Nov 18, 2006
    Messages:
    1,264
    Likes Received:
    562
    DavidGraham and BRiT like this.
  11. DavidGraham

    Veteran

    Joined:
    Dec 22, 2009
    Messages:
    3,278
    Likes Received:
    3,523
    V100 had 4 iterations: PCIe with 250w, SXM2 with 300w, SXM3 with 350w and 450w. The PCIe version delivered lower sustained performance than the SXM versions, but was not documented, this time NVIDIA is documenting this with the A100.

    https://www.anandtech.com/show/1586...lerator-250w-ampere-in-a-standard-form-factor

    The drop is greater in Multi GPU workloads though:

    https://www.tomshardware.com/news/nvidia-drops-its-first-pcie-gen4-gpu-but-its-not-for-gamers
     
    #311 DavidGraham, Jun 22, 2020
    Last edited: Jun 22, 2020
    nnunn and PSman1700 like this.
  12. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,070
    Likes Received:
    2,942
    Location:
    Finland
    PSman1700 likes this.
  13. jayco

    Veteran Regular

    Joined:
    Nov 18, 2006
    Messages:
    1,264
    Likes Received:
    562
    Yep, it was so important that then they decided to remove (well, not even bother to add) that number for the official spec list. NVIDIA does marketing and sells a lot of bullshit they days they announce new hardware, but they know it's not an absolute measure of performance, at least not for gaming.

    I was checking and there is not a TF number for even the pascal GPUs. I wonder if NVIDIA decided to stop giving meaningless numbers after the RSX announcement.
     
  14. Malo

    Malo Yak Mechanicum
    Legend Veteran Subscriber

    Joined:
    Feb 9, 2002
    Messages:
    7,614
    Likes Received:
    3,677
    Location:
    Pennsylvania
    Yeah, because the RTX launch definitely didn't see any meaningless numbers on spec sheets... :nope:
     
    jayco likes this.
  15. jayco

    Veteran Regular

    Joined:
    Nov 18, 2006
    Messages:
    1,264
    Likes Received:
    562
    it just works!
     
  16. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,070
    Likes Received:
    2,942
    Location:
    Finland
    Ampere hasn't fared any better so far, giving big fat numbers which are really big fat lies (like claiming 20x FP32 performance, not telling you that actually they're comparing FP32 of one card to TF32 of Ampere, and others in the similar fashion)
     
  17. troyan

    Newcomer

    Joined:
    Sep 1, 2015
    Messages:
    151
    Likes Received:
    241
  18. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,070
    Likes Received:
    2,942
    Location:
    Finland
    Should have been more specific, "in their slides", like this:
    https://www.computerbase.de/2020-06/nvidia-ampere-a100-pcie-karte-server/
    [​IMG]

    That slide doesn't even have fineprint to remind you to check details somewhere else
     
    #318 Kaotik, Jun 23, 2020
    Last edited: Jun 23, 2020
    Voxilla likes this.
  19. pharma

    Veteran Regular

    Joined:
    Mar 29, 2004
    Messages:
    3,528
    Likes Received:
    2,215
    Exactly where do they claim a 20x FP32 performance increase when comparing between Ampere and Volta?
    https://developer.nvidia.com/blog/nvidia-ampere-architecture-in-depth/

    Other reviews had no problems interpreting the graphs along with what Nvidia stated.
    https://www.hpcwire.com/2020/05/14/nvidias-ampere-a100-gpu-2-5x-the-hpc-20x-the-training/
     
    PSman1700 likes this.
  20. troyan

    Newcomer

    Joined:
    Sep 1, 2015
    Messages:
    151
    Likes Received:
    241
    That's a slide from the press deck. And i'm reading FP32 training. TF32 is nVidia's default precision for Training using A100.
     
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...