• Disaster@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    18
    ·
    5 days ago

    Give me something like Talos2 with a full OSS firmware and a performant CPU… and hell, a half-competitive open source graphics core too. It doesn’t need to be peak performance, it needs to be good enough.

    I’ve been trying to work with SBC’s for a while for video decoding platforms and just wound up getting stuck on x86 because the ARM situation with weirdo custom kernels for anything useful is just… annoying.

    • melpomenesclevage@lemmy.dbzer0.com
      link
      fedilink
      English
      arrow-up
      11
      arrow-down
      2
      ·
      edit-2
      5 days ago

      so, I don’t know if the shit hole made anything WORTH copying, but why respect american intellectual property? you know americans don’t respect yours. copy NVIDIA’s CUDA shit, if that’s efficient. fuck em.

      • Coriza@lemmy.world
        link
        fedilink
        English
        arrow-up
        5
        ·
        4 days ago

        The efficiency is not on the API it is on the microarchitecture. The value of copying the API is just to run unmodified software made for CUDA.

        • barsoap@lemm.ee
          link
          fedilink
          English
          arrow-up
          1
          ·
          4 days ago

          There’s also a lot of efficiency in hardware-specific kernels. A generic rocm build vs. one with hand-written kernels (not even for the proper card just a close enough one to have the same instructions) is like a 10x performance drop. That’s on the matrix multiply up to convolve these tensors level, on the layer above that you then have things like smart memory management and scheduling as well as minimising how much work needs to be done in the first place (re-ordering operations so tensors stay small) and stuff.

          You can port cuda code to vulkan or opencl – but you’re going to have to reimplement all of that. Just getting the BLAS layer to not suck is a challenge.

    • wewbull@feddit.uk
      link
      fedilink
      English
      arrow-up
      70
      ·
      6 days ago

      Yes, yes and yes, but it’ll take a while. It’s a six year project overall.

    • cocolowlander@feddit.nl
      link
      fedilink
      English
      arrow-up
      63
      ·
      6 days ago

      Feasible, yes. Practical, hard to say. Good idea, yes.

      RISC-V is open-source architecture based in Switzerland (although it started in University of California).

      One thing going for it is China is spending billions a year towards RISC-V adoption so they do not get sanctioned by the US. You need money and engineers working on it towards these type of open source to compete with existing players.

    • neon_nova@lemmy.dbzer0.com
      link
      fedilink
      English
      arrow-up
      28
      ·
      6 days ago

      With tariffs and sanctions, it has become clear that open standards which can’t be controlled by governments are what is needed.

      With what’s been happening over the past few years, there will be a lot of interested in this. Recently, I’ve seen lots of news about it, but that could just be the algorithm.

      • TheTechnician27@lemmy.world
        link
        fedilink
        English
        arrow-up
        68
        arrow-down
        1
        ·
        6 days ago

        The great thing about RISC-V if you care about sovereignty in an age where CPUs run the world is that it’s an open standard. Contrast this with x86 which is owned in some part by US-based Intel and some part by US-based AMD as well as ARM which is owned by Japanese-owned, UK-based Arm Holdings. If you want to use x86, you’re shelling out license money to Intel and AMD, and if you want to use ARM, you’re shelling out license money to Arm Holdings. You never truly “own” what you’re producing.

  • xye@lemm.ee
    link
    fedilink
    English
    arrow-up
    33
    ·
    6 days ago

    Regardless of the outcome I just hope this doesn’t lead to more tribalism in software again. The FOSS community needs to stay strong on an international level whenever it comes to hardware integration etc.

    • Tja@programming.dev
      link
      fedilink
      English
      arrow-up
      6
      arrow-down
      1
      ·
      5 days ago

      I’ll contact the maintainers of all my favorite FOSS programs written in x86 assembler, to ask them to port the software to RISC-V.

  • jim3692@discuss.online
    link
    fedilink
    English
    arrow-up
    12
    ·
    5 days ago

    There is no reference to it, but most semiconductors-making equipment is manufactured by a Dutch company named ASML. However, I don’t know how useful this will be for EU to transition to RISC-V.

    • sevenOfKnives@lemmy.blahaj.zone
      link
      fedilink
      English
      arrow-up
      19
      ·
      5 days ago

      it’s complicated. afaik asml has agreements with the us govt, and cross licensing with american companies. also, asml only makes lithography tools, there’s a LOT more to making semiconductors than just exposing patterns. and a few of the biggest vendors like kla and amat are american. kla in particular is essentially a monopoly in the metrology space.

      • melpomenesclevage@lemmy.dbzer0.com
        link
        fedilink
        English
        arrow-up
        6
        arrow-down
        2
        ·
        5 days ago

        agreements with americans are worth nothing. why keep your promises to them? they will not keep theirs to you. this whole shit show started because americans do not keep promises.

    • fenrasulfr@lemmy.world
      link
      fedilink
      English
      arrow-up
      11
      ·
      5 days ago

      It is a move to decouple from the USA for critical infrastructure. They don’t want to be in a similar situation as Ukraine in any potential conflict. Where the USA just says we will no longer allow you to use our computer chips for war with Russia.

    • qqq@lemmy.world
      link
      fedilink
      English
      arrow-up
      3
      ·
      edit-2
      6 days ago

      What’s the give away there? Not doubting just wondering.

      I see impedance matched traces so seems like something fast, but that’s all I’d be able to guess.

    • sevenOfKnives@lemmy.blahaj.zone
      link
      fedilink
      English
      arrow-up
      4
      ·
      5 days ago

      afaik, risc and cisc are pretty much the same anymore. x86, risc v and arm all have bloated instructions sets, and they all decode to risc microcode under the hood anyways.