I dunno when it happened but I swear SBCs were the new best thing in the universe for a while and everyone was building cool little servers with their RockPis and OrangePis.

Now it’s all gone x86 and Proxmox with everyone shitting on Arm. What happened? What gives?

Is my small army of xPis pointless? What about my 2 Edge routers?

I’ve got about 6 xPis scattered round my flat - is there anything worth doing with them or should I just bin them?

All thoughts, feelings and information welcome. Thank you.

  • Linkerbaan@lemmy.world
    link
    fedilink
    English
    arrow-up
    4
    arrow-down
    1
    ·
    11 months ago

    The Pi 5 isn’t very power efficient. X86 CPU’s from a few years ago were already on a more efficient process node

    • MalReynolds@lemmy.world
      link
      fedilink
      English
      arrow-up
      3
      ·
      11 months ago

      You’re quite right about the Pi 5 power efficiency, an Alder Lake N100 / i3 will smoke it in ops / watt given the right board, but the context was ‘a several year old used desktop’ which the Pi will handily beat.

      • Linkerbaan@lemmy.world
        link
        fedilink
        English
        arrow-up
        1
        arrow-down
        1
        ·
        edit-2
        11 months ago

        Depends on what it’s doing. The Pi5 has lower idle power usage but if it’s under constant load it’s actually very inefficient. Keep in mind that the Pi5 has a 25W max TDP, almost as high as the N100.

        The reason that the N100 is seems less efficient in Jeff’s video is because it’s clocked a lot higher. And power usage increases exponentially with higher clockspeeds

        The Pi5 is made on the 28nm node, which is from around 2011. Of course it has other efficiency improvements like the GPU and the ARM architecture, but pound for pound I don’t think the Pi5 even beats a 6 year old desktop in efficiency if the desktop was properly downclocked and not running some inefficient HDD’s or the likes.

        Rockchip boards on the other hand are made on 22nm, which is why they tend to be a bit more efficient.