Nvidia Ampere Discussion [2020-05-14]

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

  1. pharma

    Veteran Regular

    Joined:
    Mar 29, 2004
    Messages:
    3,582
    Likes Received:
    2,309
    Yes. From the post seems to currently used only for the ray tracing calculation portion of the benchmark.
     
    Lightman and PSman1700 like this.
  2. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    10,608
    Likes Received:
    664
    Location:
    New York
    I'm pretty sure the official benchmark doesn't support RTX. Just take a look at the scores with RTX enabled. It's a different build using different scenes but all scores are relative to a 980 so somewhat normalized.

    The 43% speedup claimed in the tweet perfectly matches up with the 2080Ti RTX off score.

    A100 4.46x (44% faster than 2080 Ti w/ RTX off)
    2080 Ti (RTX off) 3.09x
    2080 Ti (RTX on) 8.63x (93% faster than A100)

    [​IMG]
     
    #382 trinibwoy, Jul 25, 2020
    Last edited: Jul 25, 2020
    Dictator, dorf, BRiT and 2 others like this.
  3. pharma

    Veteran Regular

    Joined:
    Mar 29, 2004
    Messages:
    3,582
    Likes Received:
    2,309
    That makes more sense from an RT acceleration effect. What confused me is support for RT acceleration was included in April for OctaneRender 2020.
     
    BRiT likes this.
  4. arandomguy

    Joined:
    Jul 27, 2020
    Messages:
    1
    Likes Received:
    3
    The benchmark itself does support RT acceleration with RTX GPUs, the official score list just doesn't include those results.

    https://render.otoy.com/forum/viewtopic.php?f=7&t=70705

    Initial support was via VulkanRT about a year prior in preview builds but switched to Optix 7 which is what that release is referring to.

    https://render.otoy.com/forum/viewtopic.php?f=36&t=73237

    Previously -

    https://render.otoy.com/forum/viewtopic.php?f=7&t=71067

     
    Dictator, pharma and trinibwoy like this.
  5. pharma

    Veteran Regular

    Joined:
    Mar 29, 2004
    Messages:
    3,582
    Likes Received:
    2,309
    NVIDIA Breaks 16 AI Performance Records in Latest MLPerf Benchmarks
    [​IMG]

    [​IMG]

    [​IMG]
    https://blogs.nvidia.com/blog/2020/07/29/mlperf-training-benchmark-records/
     
    #385 pharma, Jul 29, 2020
    Last edited: Jul 29, 2020
  6. CarstenS

    Veteran Subscriber

    Joined:
    May 31, 2002
    Messages:
    4,944
    Likes Received:
    2,288
    Location:
    Germany
    "commercially available" is the key. Looking under "Research, development, or internal" one can see some very impressive numbers from Googles TPU v4, which they snuck in there.
     
  7. pharma

    Veteran Regular

    Joined:
    Mar 29, 2004
    Messages:
    3,582
    Likes Received:
    2,309
    Google set records in 6 of 8 benchmarks using it's new ML supercomputer (TPU v3). Nvidia's A100 had the fastest performance per accelerator in 8 of 8 benchmarks.
    Some other competitors decided not to submit their scores once their scores were known.
    https://syncedreview.com/2020/07/29...results-released-google-nvidia-lead-the-race/

    Edit: Regarding TPU 4.0

    https://www.enterpriseai.news/2020/07/29/nvidia-dominates-latest-mlperf-training-benchmark-results/
     
    #387 pharma, Jul 29, 2020
    Last edited: Jul 30, 2020
    PSman1700 likes this.
  8. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    10,608
    Likes Received:
    664
    Location:
    New York
  9. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,144
    Likes Received:
    3,043
    Location:
    Finland
    21 days from now is August 31st. NVIDIA revealed original GeForce August 31st in 1999, 21 years ago (though the cards actually came for sale over a month later, October 11th)
     
  10. eastmen

    Legend Subscriber

    Joined:
    Mar 17, 2008
    Messages:
    11,217
    Likes Received:
    2,252
    Look don't get me started about the geforce. I remember getting that and then the ddr versions came out and destroyed that card. Was a very very sad teenager who spent all summer cutting grass to get that card ... also damn I'm old
     
    Daozang and Rootax like this.
  11. DavidGraham

    Veteran

    Joined:
    Dec 22, 2009
    Messages:
    3,290
    Likes Received:
    3,553
    Drawing comparisons to the original GeForce implies significant performance or feature advancements.
     
    Ivan, PSman1700, pharma and 1 other person like this.
  12. Malo

    Malo Yak Mechanicum
    Legend Veteran Subscriber

    Joined:
    Feb 9, 2002
    Messages:
    7,700
    Likes Received:
    3,785
    Location:
    Pennsylvania
    One billion % performance increase! Perfect for a WCCFTech article!
     
    Daozang, CarstenS, Cuthalu and 3 others like this.
  13. Bondrewd

    Regular Newcomer

    Joined:
    Sep 16, 2017
    Messages:
    771
    Likes Received:
    361
    And 3 more YouTube videos.
     
    Malo likes this.
  14. Kyyla

    Veteran

    Joined:
    Jul 2, 2003
    Messages:
    1,032
    Likes Received:
    318
    Location:
    Finland
    Transform & Lightning?
     
    Daozang, pharma and Krteq like this.
  15. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    10,608
    Likes Received:
    664
    Location:
    New York
    Or it could be an appeal for patience while half the chip's hardware remains idle. Just like the original Geforce :)

    https://www.anandtech.com/show/391
     
    pharma and Silent_Buddha like this.
  16. PSman1700

    Veteran Newcomer

    Joined:
    Mar 22, 2019
    Messages:
    2,739
    Likes Received:
    923
    They could just launch a Turing on 7nm and it would suffice lol.
     
  17. CarstenS

    Veteran Subscriber

    Joined:
    May 31, 2002
    Messages:
    4,944
    Likes Received:
    2,288
    Location:
    Germany
    Hard to imagine that nowadays people complain about the comparatively smaller performance jumps from gen to gen.

    With today's transistor density, you could make (per CEO math) a whole GF256 (or R100) with a die size of less than 1 mm².
     
    #397 CarstenS, Aug 11, 2020 at 6:54 AM
    Last edited: Aug 11, 2020 at 7:00 AM
    BRiT and nnunn like this.
  18. Silent_Buddha

    Legend

    Joined:
    Mar 13, 2007
    Messages:
    17,098
    Likes Received:
    6,457
    I remember that card being such a disappointment. Nice demos that came with it, but uninspired in games at the time with T&L largely being unsupported. It didn't last long in my machine.

    Regards,
    SB
     
    CarstenS likes this.
  19. Rootax

    Veteran Newcomer

    Joined:
    Jan 2, 2006
    Messages:
    1,526
    Likes Received:
    883
    Location:
    France
    Well it was still fast. For me, a disappointment was the Matrox Parhelia...
     
  20. eastmen

    Legend Subscriber

    Joined:
    Mar 17, 2008
    Messages:
    11,217
    Likes Received:
    2,252
    If i recall it was a matter of months between launches but the ddr edition was not really known of before. There was no tnt sdr / ddr editions. You bought a tnt and that was it lol. Same with my riva 128 or my voodoo add in card. At least thats my view from when I was in highschool lol
     
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...