1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

News Developers urge patience over Nvidia RTX performance concerns

Discussion in 'Article Discussion' started by bit-tech, 22 Aug 2018.

  1. bit-tech

    bit-tech Supreme Overlord Lover of bit-tech Administrator

    Joined:
    12 Mar 2001
    Posts:
    3,676
    Likes Received:
    138
    Read more
     
    bit-tech, 22 Aug 2018
    #1
  2. MLyons

    MLyons 70% Dev, 30% Doge. DevDoge. Software Dev @ Corsair Lover of bit-tech Administrator Super Moderator Moderator

    Joined:
    3 Mar 2017
    Posts:
    4,196
    Likes Received:
    2,781
    The company's going in hard, too, with partnerships brown envelopes stuffed with cash that have seen a wealth of games pledge support for the new features of Nvidia's Turing architecture. FTFY. I wonder how long after the release the ray tracing support will be added.
     
    MLyons, 22 Aug 2018
    #2
  3. perplekks45

    perplekks45 LIKE AN ANIMAL!

    Joined:
    9 May 2004
    Posts:
    7,552
    Likes Received:
    1,791
    Already making excuses. RTX is off to a flying start, eh?
     
    perplekks45, 22 Aug 2018
    #3
  4. l3v1ck

    l3v1ck Fueling the world, one oil well at a time.

    Joined:
    23 Apr 2009
    Posts:
    12,956
    Likes Received:
    17
    Does that mean no extra performance over the previous cards in current games? Surely not.
     
    l3v1ck, 22 Aug 2018
    #4
  5. Gareth Halfacree

    Gareth Halfacree WIIGII! Lover of bit-tech Administrator Super Moderator Moderator

    Joined:
    4 Dec 2007
    Posts:
    17,131
    Likes Received:
    6,725
    It's entirely possible the cores are more efficient in some way, which would allow them to outperform current-gen cards even with fewer physical cores - we won't know until we see benchmarks. What we *do* know is that a GeForce GTX 2080 Ti with all CUDA cores would outperform the GeForce RTX 2080 Ti with its mixture of CUDA, RT, and Tensor cores for games which do not make use of the RT or Tensor cores (which, at the moment, is all of them.)
     
    Gareth Halfacree, 22 Aug 2018
    #5
  6. edzieba

    edzieba Virtual Realist

    Joined:
    14 Jan 2009
    Posts:
    3,909
    Likes Received:
    591
    If you normalise to the same die area (and same manufacturing cost), that is. And assuming that doesn't result in a bottleneck somewhere else (e.g. memory bandwidth to feed all those CUDA cores).
     
    edzieba, 22 Aug 2018
    #6
  7. perplekks45

    perplekks45 LIKE AN ANIMAL!

    Joined:
    9 May 2004
    Posts:
    7,552
    Likes Received:
    1,791
    Cost has no impact here. The same die area is a given, but filled to the brim with "traditional" CUDA cores instead of the massive area reserved for ray tracing. See here, the green block is just ray tracing:

    [​IMG]

    Bandwidth should not be a bottleneck, what with a 352 bit bus and 14 Gbps GDDR6.
     
    perplekks45, 22 Aug 2018
    #7
  8. Anfield

    Anfield Multimodder

    Joined:
    15 Jan 2010
    Posts:
    7,062
    Likes Received:
    970
    Not sure we can make that assumption since Raytracing may or may not carry significantly higher requirements.
    But yes, for 99.9% of games it won't be an issue.
     
    Anfield, 22 Aug 2018
    #8
  9. perplekks45

    perplekks45 LIKE AN ANIMAL!

    Joined:
    9 May 2004
    Posts:
    7,552
    Likes Received:
    1,791
    Well, we were talking about theoretically replacing all RT cores with CUDA cores, so we are in known territory for bandwidth.
     
    perplekks45, 22 Aug 2018
    #9
  10. edzieba

    edzieba Virtual Realist

    Joined:
    14 Jan 2009
    Posts:
    3,909
    Likes Received:
    591
    That's a Pretty Slide Picture rather than an actual die layout diagram.
    If you look at the die shot (assuming even that is accurate) you can see 144 SMs for the fully operational die. Looking at the Quadro RTX 8000, that means 32 CUDA cores per SM (to make up 4608), 4 Tensor cores per SM (to make up 576) and an unknown number of RT cores .
    ::EDIT:: Herped a derp, SMs are mirrored. 72 SMs, 64 CUDA cores per SM, 8 Tensor cores per SM. A little birdie also says 72 RT cores, but not whether that is 1 per SM as part of the SM or a separate block independent of the SMs (I'd guess as part of the SM myself).

    If we go by previous generation die layout breakdowns, the central 'column' is composed of the main uncore controller in the centre, the 'top' and 'bottom' of the bands being the ingest pipelines. Around the periphery are the memory controllers and ROP units (combined).
     
    Last edited: 23 Aug 2018
    edzieba, 23 Aug 2018
    #10
  11. tozsam

    tozsam What's a Dremel?

    Joined:
    11 Jun 2010
    Posts:
    87
    Likes Received:
    0
    "Windows, Xbox One, and PlayStation 4, with only the former receiving ray tracing support" Wouldn't this be to do with consoles not having RT cores??? Shouldn't be a suprise
     
    tozsam, 5 Sep 2018
    #11
Tags: Add Tags

Share This Page