Nvidia Ampere Discussion [2020-05-14]

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

Tags:
  1. sonen

    Newcomer

    Joined:
    Jul 13, 2012
    Messages:
    80
    Likes Received:
    102
  2. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    11,467
    Likes Received:
    2,271
    Location:
    New York
    Nvidia claims Ampere already supports “Smart Access Memory” and says they’re working on implementing it in drivers.

     
  3. Krteq

    Newcomer

    Joined:
    May 5, 2020
    Messages:
    80
    Likes Received:
    138
    Every IHV supports this. All you need to have is a support of that optional feature in HW (PCIe controller) and SW :)
     
    pharma and Lightman like this.
  4. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    11,467
    Likes Received:
    2,271
    Location:
    New York
    The burning question is why is it only being enabled now?
     
    Lightman likes this.
  5. pharma

    Veteran Regular

    Joined:
    Mar 29, 2004
    Messages:
    4,397
    Likes Received:
    3,737
    Wonder if they will "trickle down" support to non-Ampere cards? Since my X-99 MB has a toggle to enable it, seems not to require pcie-4 or Ampere cards.
     
  6. SimBy

    Regular Newcomer

    Joined:
    Jun 21, 2008
    Messages:
    700
    Likes Received:
    391
    Probably not a whole lot of real world benefit before, with lesser anounts of VRAM? But most importantly, AMD wasn’t lighting a fire under Nvidias rear. Now every frame counts.
     
  7. Man from Atlantis

    Regular

    Joined:
    Jul 31, 2010
    Messages:
    928
    Likes Received:
    767
    I could see this might be happening because of a public backlash when AMD announced no Zen3 support for X470/B450 boards.
     
    PSman1700, Lightman and pharma like this.
  8. pharma

    Veteran Regular

    Joined:
    Mar 29, 2004
    Messages:
    4,397
    Likes Received:
    3,737
    Curious if the benefits from using a feature like SAM would be included if talking about IPC gains, or is considered a separate metric?
     
  9. pjbliverpool

    pjbliverpool B3D Scallywag
    Legend

    Joined:
    May 8, 2005
    Messages:
    8,844
    Likes Received:
    3,510
    Location:
    Guess...
    That's great news, hopefully this becomes more widespread and we see some really gains from it. Once it becomes the standard rather than the exception game developers can start targeting it and in some ways the PC with its separate components can take advantage of something akin to a UMA.

    Does anyone know if GPU's can read directly from system memory in the same way? I *think* that's what HBCC allowed but you had to manually set a potion of the system memory to be used as VRAM.
     
    PSman1700 likes this.
  10. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    11,467
    Likes Received:
    2,271
    Location:
    New York
    That doesn’t explain why AMD didn’t enable it earlier. They certainly needed the performance.
     
    PSman1700 and Lightman like this.
  11. Scott_Arm

    Legend

    Joined:
    Jun 16, 2004
    Messages:
    14,853
    Likes Received:
    7,120
    Nvidia has only mentioned enabling it with Ampere, not Turing. Seems like maybe neither of them had implemented it until this new generation.
     
    Lightman likes this.
  12. OlegSH

    Regular Newcomer

    Joined:
    Jan 10, 2010
    Messages:
    663
    Likes Received:
    1,242
    To me, it looks like they've already enabled it on driver level for DX12 games on current GPUs.
    It would be weird not to do it since host addressable video memory has been available in Vulkan for years and both NVIDIA and AMD supported it for quite a while (256 MBs though, but this might be enought to store hot descriptors/constants/other stuff which changes on frame to frame basis and is hard to cache on GPU, especially with some devs' love).
    I see RX 5700 XT being able to compete with RTX 2080 in some recent titles like Valhalla in 1440p and below, while still being much slower in 4K and it's not the only example.
    It would be nice if somebody tested this title with the most recent driver vs year old driver so that we can verify whether there is something in recent drivers, which can boost performance a lot in titles like Valhalla.
     
  13. OlegSH

    Regular Newcomer

    Joined:
    Jan 10, 2010
    Messages:
    663
    Likes Received:
    1,242
    That's exactly how it's done today, command processor in GPU reads everything from host memory.
    In Vulkan, a dev can push constants/etc to GPU's video memory so that it's faster to read them on GPU == less overhead due to latencies when there are tons of descriptors/constants/etc
     
    pharma, PSman1700 and pjbliverpool like this.
  14. Rootax

    Veteran Newcomer

    Joined:
    Jan 2, 2006
    Messages:
    2,236
    Likes Received:
    1,686
    Location:
    France
    Maybe not worth the driver work needed ? I mean, some games are not 100% stable with HBCC on, while perfectly fine withtout it (ffxv come in mind for me). So I guess they didn't want to check if this was working ok with older gpus ?
     
  15. Scott_Arm

    Legend

    Joined:
    Jun 16, 2004
    Messages:
    14,853
    Likes Received:
    7,120
    sebbbi talks resizable RBAR support and why it's good.

    I wonder if Nvidia has planned this as part of their DirectStorage support and are now pulling this particular feature forward so that they can get some early gains in the benchmark wars.

     
    T2098, Rootax, pharma and 5 others like this.
  16. Kaotik

    Kaotik Drunk Member
    Legend

    Joined:
    Apr 16, 2003
    Messages:
    9,960
    Likes Received:
    4,144
    Location:
    Finland
    That 256MB is always supported AFAIK no matter what API or GPU, the difference is that now that 256MB can be turned into however many gigabytes there is VRAM on each particular card
     
  17. Frenetic Pony

    Regular Newcomer

    Joined:
    Nov 12, 2011
    Messages:
    783
    Likes Received:
    454
    Sebbi's been on a tear lately with the advice and explanations. Love it and love seeing RBAR as an open standard as compared to smart access memory. Really need one standard across all platforms, hopefully looking as much like consoles as possible as well, to reduce the amount of individual platform support that's needed.
     
    Lightman and pjbliverpool like this.
  18. Jawed

    Legend

    Joined:
    Oct 2, 2004
    Messages:
    11,377
    Likes Received:
    1,687
    Location:
    London
    Why do you think SAM is not RBAR?
     
    Frenetic Pony, pharma and SimBy like this.
  19. SimBy

    Regular Newcomer

    Joined:
    Jun 21, 2008
    Messages:
    700
    Likes Received:
    391
    You would think at least more tech savvy people on this forum knew that SAM is just how AMD calls it's RBAR implementation on Windows platform. I mean there was literally a video posted right after announcement where Frank Azor explains in detail how it works.
     
  20. trinibwoy

    trinibwoy Meh
    Legend

    Joined:
    Mar 17, 2004
    Messages:
    11,467
    Likes Received:
    2,271
    Location:
    New York
    If Nvidia can confuse people with RTX it’s only fair that AMD can do the same with SAM.
     
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...