My computer has been having a weird problem I'm not sure how to diagnose, and I'd really appreciate any help narrowing this down
Symptom: the GPU fans spin up to max, the screens go black, and the computer becomes impossible to interact with. My CPU cooler has a temperature display on it, and it also goes blank. Audio will still work; I've had this happen when I was on a discord call, and I continued to be able to talk to the people in this state. This state will continue indefinitely until I hold the power button, after which it will shut down. If I try to power it up, the fans will twitch slightly, but nothing else happens.
Frequency: this happens every few days (doesn't seem to matter what I'm doing on the computer), and almost always if I try to restart my computer through Windows
My workaround: I've found, through trial and error, if I unplug the computer, remove the CMOS battery, hold the power button for a minute, leave the computer alone for a few hours, plug it back in, and flash the BIOS, it will work for a few days. This issue used to happen once ever few months, but has increased in frequency over the last year to where it happens every ~3 days.
Things I've tried:
- Reliability monitor - tells me this is a GPU issue with the GPU getting removed
- HWiNFO64 - I've caught an event like this in those logs. Prior to the event, nothing is obviously wrong; the GPU power and temperature remain stable, and then the GPU temp just drops to 0 as the GPU disconnects. I've also seen the "PCIe PEX Errors Recovery Counter" growing, into the thousands
- All the things in this thread. Did not make a difference
- Replacing my CMOS battery
- Updating all my drivers
My specs:
- GeForce RTX 4080 SUPER
- X670E Pro RS
I've been (unsuccessfully) trying to solve this issue myself, as you can see - but I'm all out of ideas. Does anyone know what could be causing this or how I can fix it?