• udc@lemmy.world
    link
    fedilink
    English
    arrow-up
    8
    ·
    1 day ago

    How close are we to RISCV chips having comparable performance to consumer proprietary chips?

  • navordar@lemmy.ml
    link
    fedilink
    English
    arrow-up
    12
    ·
    edit-2
    23 hours ago

    It’s beyond my mind that x86(-64) survived to this day. It’s so inefficient

    Edit: to clarify: I know nearly nothing about processors, so maybe the architecture isn’t really that inefficient. All I know is that after Apple switched to its own ARM-based chips, the programs could run faster using the same or less energy. Even with the compatibility later

      • Draconic NEO@lemmy.dbzer0.com
        link
        fedilink
        English
        arrow-up
        4
        ·
        22 hours ago

        This is one that way too many people forget about, to switch to something else, you’d need to emulate x86 for the software that hasn’t or do without. Emulation causes overhead and performance reduction, and doing without seems fine on paper until you encounter software you want or more crucially need and can’t use it because it hasn’t been ported, isn’t open-source, or can’t be ported due to heavy use of x86 assembly.

        • _MusicJunkie@beehaw.org
          link
          fedilink
          English
          arrow-up
          2
          ·
          21 hours ago

          On the other hand, nobody ever will fix, update, or port many pieces of software if backwards compatibility is expected to be kept up indefinitely.

          I’m frankly not sure which is better. Do it the apple way, force projects to make their software work on modern platforms, and just accept that some people will really be fucked over for the sake of progress. Doesn’t sit right with me, but on the other hand things can’t keep being compatible forever.

    • easily3667@lemmus.org
      link
      fedilink
      English
      arrow-up
      6
      ·
      1 day ago

      It’s efficient on a macro scale. Imagine if every piece of software on the planet had to be recompiled and tested for each flavor of the month risc processor?

  • resipsaloquitur@lemm.ee
    link
    fedilink
    English
    arrow-up
    17
    ·
    edit-2
    2 days ago

    I’m all for open ISAs, but one of these can be licensed at reasonable rates and the other sued AMD for making the 64-but instruction set architecture that Intel should have made.

  • Godort@lemm.ee
    link
    fedilink
    English
    arrow-up
    33
    ·
    2 days ago

    I’m fully ready to switch to RISCV when compatibility and performance match x86_64 and ARM