• olgas_husband
      link
      fedilink
      arrow-up
      9
      ·
      11 months ago

      I’ve seen this shit, it gets me that this dude is cynical to the point that he says that the card is new and the drivers aren’t fully opengl and directx compliant yet, it is focused on ia and at the end spill the old oriental cheap knockoff fallacy

  • blashork [she/her]@hexbear.net
    link
    fedilink
    English
    arrow-up
    19
    arrow-down
    1
    ·
    11 months ago

    What really fucking gets me about this article is how they note that Moore already has a stack that can run cuda crap on it. This shit that AMD has not been able to pull their god damn heads out of their asses and do fully and completely for years, and here they already have it ready to go as they’re essentially at the beginning of the chinese homegrown gpu story. Like, this is something that’s going to be a decades long project, and they already have that critical part built. American computing is a fucking joke and should be burned to the ground.

    • trashxeos
      link
      fedilink
      English
      arrow-up
      13
      ·
      11 months ago

      Also, they’ve been doing this shit while under heavy sanctions regarding lithography tech from the EU and the USA government bodies. Imagine where they’d be if they weren’t. I’d put money that at their current rate of development, they’ll be able to achieve or exceed parity in both function and efficiency within the next decade.

      • blashork [she/her]@hexbear.net
        link
        fedilink
        English
        arrow-up
        6
        ·
        11 months ago

        I’m well aware. While I don’t personally work on the gpu related stuff, I do work at a company that has to do a lot of gpu computing. My opinions on this topic are mostly informed by coworkers who do write gpu code, specifically a lot of opencl kernels. opencl has a lot of shortcomings and issues, and the kernel model is a removed to work with. However, that isn’t the point. The point is about compatibility. You can have a really good gpu but if you don’t have adversarial compatibility with your competitors, it will just die. Specifically, amd have done a shit job at making cuda run on amd gpus. rocm is a disjointed mess, it sucked when I had to work with it in uni, it still sucks now. Cuda is bad and proprietary, but any modern gpu should still be able to run cuda crap simply because it’s useful to be able to do so, and there’s a lot of things already built with it that should remain accessible.

        The fact that amd have not been able to get a component as critical as their adversarial compatibility layer working while Moore has already implemented it for their early generation of cards shows:

        • how profound the failures of amd and western computing companies is
        • the breakneck pace and incredible technical achievements of chinese gpu development
        • الأرض ستبقى عربية
          link
          fedilink
          arrow-up
          5
          ·
          11 months ago

          It may also mean that Moore Threads has more legal freedom being Chinese to implement CUDA compatibility even if it infringes on Nvidia patents. Sometimes the limitations are legal rather than technical, either way AMD is in serious trouble if Moore Threads achieves sufficient CUDA compatibility