• secret300@lemmy.sdf.org
    link
    fedilink
    English
    arrow-up
    7
    arrow-down
    1
    ·
    10 days ago

    These companies have billion to spend on developing a chip so why give that money to arm when they could make their own risc-v chip without royalty fees

    • tekato@lemmy.world
      link
      fedilink
      English
      arrow-up
      4
      ·
      10 days ago

      This has probably been in the works for years, and RISC-V’s profile (RVA23) for proper user application support was only released a few days ago.

  • Telorand@reddthat.com
    link
    fedilink
    English
    arrow-up
    4
    arrow-down
    3
    ·
    11 days ago

    I have generally negative feelings about big companies getting bigger, but from what I understand about the difference between instruction sets, the transition to ARM is a good thing.

    • bamboo@lemm.ee
      link
      fedilink
      English
      arrow-up
      7
      arrow-down
      1
      ·
      edit-2
      11 days ago

      More competition for CPUs in laptops/desktops/servers is a good thing, especially if these can compete on efficiency.

      • Telorand@reddthat.com
        link
        fedilink
        English
        arrow-up
        1
        ·
        edit-2
        10 days ago

        The number of Assembly instructions needed to do similar things tends to be less on ARM, from my limited understanding. That leads to efficiencies.

        • pycorax@lemmy.world
          link
          fedilink
          English
          arrow-up
          7
          ·
          edit-2
          10 days ago

          That’s not really how it works actually. You got sort of the idea that ARM is a Reduced Instruction Set Computer (RISC) architecture but the reduction here refers more to the variety of instructions rather than amount of instructions. In fact ARM typically requires more instructions since there’s less varieity.

          But that really doesn’t mean much in modern processor architectures since all modern processors decode assembly instructions into micro operations internally and execute them. Each instruction and their corresponding micro operations may have a different number of cpu cycles to execute so it’s not something that’s so easily calculatable.

          The age of RISC vs CISC (x86, etc) debates has largely ended because of how modern CPUs work. The difference between instruction sets mostly just come down to the language that compilers translate to.