• Morningst4r@alien.topB
    link
    fedilink
    English
    arrow-up
    1
    ·
    10 months ago

    More cores isn’t necessarily more efficient though. A lot of SOCs would prefer to boost one or two high performance cores

    • manek101@alien.topB
      link
      fedilink
      English
      arrow-up
      1
      ·
      10 months ago

      It is kinda more efficient tho.
      Power consumption per clock doesn’t scale linearly.
      Splitting a task into two parts and running it on two cores at 2 GHz will consume less power and generate less heat than running it on a single core at 4 GHz.
      Its significantly harder to do that programming wise, but performance wise it has its benefits.
      It also frees up the main high performance core to focus on more important tasks.

      • Morningst4r@alien.topB
        link
        fedilink
        English
        arrow-up
        1
        ·
        10 months ago

        That’s assuming you can split it efficiently without any overhead. It’s not just a problem to be “solved” that devs don’t bother with, a lot of tasks scale poorly across threads even if you successfully multi thread them