1. OneDrive mirroring

Stop telling me that I am out of storage, I did not ask you to put my files there.

  1. Nvidia Linux drivers

A 2 month old driver from a desktop 3070 failed to detect a laptop 3050, and I had to chroot to remove the offending driver, buuuut

  1. Rescue USB

So it turns out the default Tumbleweed ISO is NOT a live image. Recue didn’t work nor did usb boot until I realised. Burned LMDE onto the usb key, and I was able to get back up and running relatively quickly.

  • Findom_DeLuise [she/her, they/them]@hexbear.net
    link
    fedilink
    English
    arrow-up
    4
    ·
    1 month ago
    1. When launching Microsoft-occupied Minecraft, it kills all sound to my main audio device until I change the audio sampling rate. As soon as I change the sample rate, sounds start playing as if they were queued up and waiting for the device. Exclusive mode is disabled, but it does this either way. Could be a Realtek/Nahimic driver issue, but it only ever happens with MC Bedrock Edition, and the game is fine after that until I relaunch.
    2. “Sleep mode” is more of a gentle suggestion than anything that actually puts my PC into a sleep state. There are no active wake timers, and the power event history shows fuck all. It just wakes up after 20-30 seconds for no reason sometimes. On occasion, putting it back to sleep several times will make it stop.
    3. …I can’t really think of a #3 that is software-related. I knew what I was getting into when I set this thing up with Windows 11, and I think I’ve got Microsoft’s bullshit fairly well locked down, at least until the next major update reverts all of my unenshittification settings and resurrects Copilot and Edge spyware.

    Now, hardware-wise? Let’s just say that AM5 is a strange, strange beast. Even with direct-die water cooling, temps regularly spike into the 70s just while internet browsing. Then they drop back down to high 20s/low 30s within 15-20 seconds, but dear fucking god this thing can very suddenly start cranking out copious amounts of heat. For reference, I upgraded my build from an i7 6700k rig that was also water-cooled (albeit not direct-die), and that never really got above 50 C. Anyway, this is “normal” behavior for AM5 chips (and especially the X3D series, and even more so for Ryzen 9 tier), so I guess I just have to suck it up and get used to my water loop’s fans occasionally going into Emergency Leaf Blower mode.

    • PorkrollPosadist [he/him, they/them]@hexbear.net
      link
      fedilink
      English
      arrow-up
      2
      ·
      edit-2
      1 month ago

      Even with direct-die water cooling, temps regularly spike into the 70s just while internet browsing.

      I don’t know if this is just cope, but I read somewhere that this is by design. These chips run very close to their thermal limit, because doing otherwise means you could squeeze extra performance out of it and are choosing not to. I switched from an i7-4790k to a r9-7950x3d (admittedly, a decade apart) and the thermal behavior is very different. The temps on the AMD chip shoot up very quickly, but I haven’t seen them go out of spec like I’ve seen on the Intel. The Intel would generally stay cooler, but certain tasks like video encoding or Prime95 would drive it straight up to 100C and cause a thermal halt.

      • Findom_DeLuise [she/her, they/them]@hexbear.net
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 month ago

        It is definitely by design. The current generations from both manufacturers are absurdly efficient at dissipating heat from the die. The 3D cache layer knocks off some thermal headroom, but the modern chips are smart enough to thermal throttle before they hit their rated thermal ceiling. This is part of why undervolting is so effective with the AM4 and AM5 chips. For what it’s worth, the AM4 Ryzens do not have temp spikes like this. And Intel seems to be hell-bent on making toaster ovens…

        Weird that a Haswell i7 would run that hot – unless I’m misremembering, you might have had one with the shitty thermal paste between the die and IHS; it wasn’t an issue on all Haswell and Ivy Bridge chips, but they definitely had some QC issues with the paste drying out during those generations. I also had a 4790k in a workstation machine that I tricked work into paying for, and even on just a 120mm AIO (without the delid/liquid metal treatment), it still stayed below 60-65C during thermal torture tests (Linpack and Prime95 small FFTs). Maybe I did delid it and bring in some spare LM from home; it was a long time and several dead brain cells ago.