Leaked specs for the fastest Navi 48 and entry-level Navi 44 graphics cards, in addition to performance estimates, paint a bright future.
  • John Richard
    18•8 months ago

    If it has SR-IOV in the consumer version then yes I need it. Otherwise f*ck both AMD & NVIDIA.

    • @RamblingPanda
      6•8 months ago

      I’ve been out of the loop for some years, could you eli5 for me please?

      • Transient Punk
        6•8 months ago

        SR-IOV allows you to share your GPU among many virtual machines in much the same way that you are able to share a single CPU among many VMs

        • @RamblingPanda
          1•8 months ago

          Thanks for the info. I guess that’s not that essential for gaming but more for AI specific tasks?

          • @[email protected]
            4•8 months ago

            Not really ai specifically but VMs. Maybe you want a Windows vm for gaming with a gpu, just give it a slice and it’s fine. Maybe you want lots of VMs for all various different office clients, split off sections of the gpu and you can have a bunch of hardware accelerated thin clients

            • @RamblingPanda
              2•8 months ago

              I’ve never thought about that, but it makes sense. Thanks for the explanation!

          • Transient Punk
            3•8 months ago

            For me, I would love to have a single GPU in my server that I can split up for use in transcoding videos for Plex in one VM, and another VM running something like Blue Iris with AI video analysis.

            The potential use cases are many and varied, including some gaming use cases. You could have a single GPU in your Linux desktop, and be able to pass that through to a Windows VM to get native performance gaming in a VM. This is technically already possible, but you need two GPUs. With SR-IOV you could get away with only having one

            • @RamblingPanda
              3•8 months ago

              Ok, now I want it too. Thanks for the explanation!

    • @[email protected]
      1•8 months ago

      Okay a quick search tells me this is short for Single-root input/output virtualization right? Can you explain why that would be advantageous in a GPU?

  • RiQuY
    14•8 months ago

    Hot take but what people need is ROCm support on all RNDA GPUs from 1 to 3, as they promised in the past.

    Seems wild to me that even a GTX 1050M can run CUDA but the ROCm hardware support is so limited and shitty.

    • Ricky Rigatoni
      4•8 months ago

      As someone who would prefer to make my 5700XT last the 9 years my 5770 did, I agree. I don’t want to be buying a new card every other year just because they didn’t bother continuing to support my current.