I’m in the process of getting my Home Assistant environment up and running, and decided to run a test: it turns out that my gaming PC (custom 5800X3D/7900XTX build) uses more power just sitting idle, than both of my storage freezers combined.

Background: In addition to some other things, I bought two “Eightree” brand Zigbee-compatible plugs to see how they fare. One is monitoring the power usage of both freezers on a power strip (don’t worry, it’s a heavy duty strip meant for this), and the other is measuring the usage of my entire desktop setup (including monitors and the HA server itself, a Lenovo M710q).

After monitoring these for a couple days, I decided that I will shut off my PC unless I’m actively using it. It’s not a server, but it does have WOL capability, so if I absolutely need to get into it remotely, it won’t be an issue.

Pretty fascinating stuff, and now my wife is completely on board as well; she wants to put a plug on her iMac to see what it draws, as she uses it to hold her cross-stitch files and other things.

7 months later update:

I’ve expanded HA quite a bit and have a decent grasp on where my electricity is being used. Suspending my desktop has saved a shitload of power: under Linux, when suspended, it draws ~10W of power at any given time vs ~100W when idle. Also, I put smart plugs on the server cluster; the total usage of the entire cluster - 4 mini PCs, 1 SFF, and 5x HDDs (NAS) - over the last 7 months so far is currently at a whopping 465 kWh. That’s ~66 kWh a month. Barely 5% of my monthly electrical usage.

  • aichan@piefed.blahaj.zone
    link
    fedilink
    English
    arrow-up
    2
    ·
    11 days ago

    Can I ask what smart plugs and Zigbee antenna you are using? We are about to do the same, so it would be great to get something already tested ^^

  • slazer2au@lemmy.world
    link
    fedilink
    English
    arrow-up
    1
    ·
    8 months ago

    What kind of freezers are they? I hear that top loading freezers are quite efficient because the cool doesn’t escape when it gets opened like a front loading one.

    • Lka1988@sh.itjust.worksOP
      link
      fedilink
      English
      arrow-up
      0
      ·
      edit-2
      8 months ago

      One is a smaller chest freezer, about 3 feet tall, probably 6 or 7 cubic feet if I had to guess. The other is a Hamilton Beach upright freezer from Costco. Both are full, so that helps with keeping them cold.

      • AnarchoSnowPlow@midwest.social
        link
        fedilink
        English
        arrow-up
        0
        ·
        8 months ago

        Is your upright the one with all the little compartments? That one looked to me like the most efficient upright design I’ve ever seen.

        • Lka1988@sh.itjust.worksOP
          link
          fedilink
          English
          arrow-up
          1
          ·
          edit-2
          8 months ago

          Yep, it’s awesome. We got it for $300 from Costco to supplement the smaller chest freezer, and it’s been an absolute godsend.

  • fmstrat@lemmy.nowsci.com
    link
    fedilink
    English
    arrow-up
    0
    ·
    8 months ago

    Couple of thoughts:

    1. That smart plug may not be rated to the max wattage when GPU and CPU are at full blast. Be careful, because that could be an expensive mistake. Place a surge protector between the smart plug and the PC to be safe. Also run the PC full tilt for a while and make sure the smart plug doesnt get warm. If it does, fores have been known to start from those.

    2. Sounds like you know this with WoL, but suspend is your friend 😉 If the gaming PC is linux and you run into suspend issues, let me know, I’ve seen 'em all.

    • Lka1988@sh.itjust.worksOP
      link
      fedilink
      English
      arrow-up
      1
      ·
      8 months ago

      Perfect, I don’t need to run the fans anymore!

      Seriously though - we have 5 kids, and feeding the little shits is expensive, so we freeze a lot of things for storage. I thought for certain the freezers would be power hogs compared to an idling PC, but I was very surprised to be proven wrong.

      Next up… Measuring my server cluster 😬