NVIDIA A100-Powered (PCIe) Servers to Accelerate AI, Data Science and Scientific Computing

Discussion in 'Frontpage news' started by Hilbert Hagedoorn, Jun 22, 2020.

  1. Hilbert Hagedoorn

    Hilbert Hagedoorn Don Vito Corleone Staff Member

    Messages:
    48,544
    Likes Received:
    18,856
    GPU:
    AMD | NVIDIA
    DeskStar likes this.
  2. Death_Lord

    Death_Lord Guest

    Messages:
    722
    Likes Received:
    9
    GPU:
    Aorus 1080 GTX 8GB
    I wonder if it would be posible to use one of those as RayTracing and Physics accelerators for games, GPU for the main graphics, Accelerator card for simulations and RT.
     
  3. PurSpyk!!

    PurSpyk!! Master Guru

    Messages:
    730
    Likes Received:
    37
    GPU:
    RTX 4090
    That card has not RT cores, so it would be a bit pointless?
     
  4. Denial

    Denial Ancient Guru

    Messages:
    14,207
    Likes Received:
    4,121
    GPU:
    EVGA RTX 3080
    The card doesn't have RT cores and even if it did the answer is probably no.

    Eventually GPUs will go MCM, especially for compute purposes, but the latency and scheduling complications make it near impossible to have a RT accelerator card.
     

  5. DeskStar

    DeskStar Guest

    Messages:
    1,307
    Likes Received:
    229
    GPU:
    EVGA 3080Ti/3090FTW
    Bad ass!.!
     
  6. Gomez Addams

    Gomez Addams Master Guru

    Messages:
    258
    Likes Received:
    166
    GPU:
    RTX 3090
    For this card's intended audience, raytracing cores are not needed at all. In fact, the card has no video outputs and it doesn't need any. What it does have is 108 streaming multiprocessors and 40GB of HDM2. Those are both outstanding so this card should be a monster. It's cost likely will be also.
     
  7. Astyanax

    Astyanax Ancient Guru

    Messages:
    17,040
    Likes Received:
    7,381
    GPU:
    GTX 1080ti
    no.
     

Share This Page