4080 Super crashing around once a day
Posted by ManofShadows@reddit | buildapc | View on Reddit | 6 comments
I'm having a really tough time troubleshooting this issue, but my build which has otherwise been stable for around half a year is experiencing an issue where the GPU will crash 30 mins to a couple hours after my first boot of the day. By crash, I mean that the rest of the PC remains running but the GPU feed cuts to my monitors and the fans start spinning at max RPM, until I manually power down the PC with the power button and turn it back on again.
I have seen several instances of a similar issue posted online where people have indicated it may be a power supply issue, but my build has a ASUS TUF Gaming 1000W Gold which should be more than sufficient for this GPU, and this issue has only occurred when the GPU is *not* under significant load (only a couple browser tabs open, discord, etc. no games). I have not yet encountered this issue while stressing the GPU via blender or gaming. I am on the latest GPU drivers as well, and it seems to me that the GPU power connector is as firmly attached as I can make it.
For whatever reason this issue has never occurred multiple times a day, and I usually use the GPU more for gaming in the evening, where I would expect this to be happening more.
Hoping for some guidance on troubleshooting this further as I am feeling quite stumped. I have considered maybe picking up a UPS in case it is an actual voltage drop in my home causing this? But I am not sure if that's a justifiable concern.
Razgorths@reddit
This sounds to me more like a CPU issue than GPU. I had this exact behavior when I was testing my CPU undervolt; conversely, when testing my GPU undervolt, it would always BSOD.
The logic here is that as long as your CPU is functioning, it should be able to throw a BSOD even if your GPU malfunctions; however if your CPU malfunctions there's no way to throw a BSOD because the processor...isn't functioning and can't calculate what the error was.
Check your Event Viewer for errors, especially WHEA-Logger errors around the time of crashing.
ManofShadows@reddit (OP)
I do see a number of errors logged around the time I remember the crash happening. There's a flurry of DistrubutedCOM errors all with the same ID
Searching the key in regedit shows an entry with the data "Thumbnail Cache Out of Proc Server"
There's also a number of application errors indicating the desktop windows manager crashing repeatably around that time. I'm not sure if that could be the cause or potentially just a symptom of the actual problem? That error looks like this:
As for the CPU, I *think* it remains functioning. At least, the last time this happened earlier today, I had media playing at the time, and the media continued to play even for \~30 seconds after the issue occurred.
Razgorths@reddit
Lots of third-party data on dwmcore errors saying it's related to either GPU drivers or Xbox Game Bar or GPU power connectors etc, I would Google around and try some of those fixes first.
Stargate_1@reddit
I do not think this clearly indicates a GPU issue. You could do that /sfc scannow command and see if any corruption in your system comes up, i dont remember the exact command tho
ManofShadows@reddit (OP)
I ran this and it did say that it found & fix corrupted files, if this actually was the issue I am going to be a very happy windows user!
```
Beginning system scan. This process will take some time.
Beginning verification phase of system scan.
Verification 100% complete.
Windows Resource Protection found corrupt files and successfully repaired them.
For online repairs, details are included in the CBS log file located at
windir\Logs\CBS\CBS.log. For example C:\Windows\Logs\CBS\CBS.log. For offline
repairs, details are included in the log file provided by the /OFFLOGFILE flag.
```
PrisonerV@reddit
Use afterburner to reset default settings on the video card, especially fan speed.