The only critical event at the time of the crash in Event Viewer Administrative Events view is an Event ID 41 (Kernel-Power), but I don't know if there's anything else I should be looking for.
Comments
Log in with your Bluesky account to leave a comment
video card power issue? once the video card starts working hard it browns out, reseat any power cable going to the card on both ends, try a better cable, some cables don let enough power in (card needs 300w cable only allow 150 through?) or power supply (the one you listed seems ok on paper tho)
Check seating of CPU power connector (less likely to be this)
Check seating of CPU fan power connector, and ensure that it's in the right header on the motherboard. If it's not in the PWM header, your CPU might not be responding correctly to increased load.
It's at least somewhat noticeable that the games that are crashing are ones where I'd expect a higher CPU load. Failure of the fans to respond to thermal load (as suggested by steadily increasing temps) can cause Kernel41 errors.
We've tried to RMA it. Manufacturer says they couldn't find a problem with the card, but they obviously can't be trusted. Local computer shop also couldn't find a problem with the card; they say they've been running it under load using benchmarking software for hours and there are no issues.
I don't have enough knowledge of computers to evaluate whether the amount of paste on my CPU is the right amount. The amount I put on there is consistent with the assembly instructions I had.
Ok, from what you describe it could be a temperature issue. Have you tried running any kind of GPU 3dmark benchmark software to see if you can provoke the crash?
There's also some dedicated CPU and memory test tools you could try such as OCCT.
2.) The buzzing reeks of a conflict on hardware level; can happen if the GPU is not plugged into the faste PCIe slot (the one next to the CPU is the correct one).
It's in the fastest PCIe slot but it's a heavy GPU and even though we only had it plugged in for long enough to notice before putting it on its side and adding a support, I wouldn't be surprised if it fucked the port
So, did it change for the better when you supported the GPU?
Thing is, if it's not the GPU (surprising, given that this problem occurs as soon as videogames are being started up) then I would check the system RAM next. It's unlikely though, because the system runs in the first place.
It sounds like an overheating issue. It might sound stupid and crude, but try putting a box fan next to it with the case open and see if it crashes still.
Sounds like a terrible time, good luck on fixing it! (When my custom pc started crashin it was cause the actual drivers were missin a digital signature, i ended up having to reinstall windows 11 and has a friend mess with the drivers)
You're playing a few games there that are very high performance, the problem is Windows, especially since 11, has so much bloat it's a never ending slog to disable background processes if youre doing anything intensive.
Games with which this has occurred so far include:
ELDEN RING (2022)
CYBERPUNK 2077 (2021)
STAR WARS — JEDI: FALLEN ORDER (2019)
Games with which this hasn't occurred so far include:
THE FORGOTTEN CITY (2021)
THE ELDER SCROLLS V: SKYRIM — Special Edition
THE ELDER SCROLLS ONLINE (2014)
Solutions which have been suggested and tried so far:
1. Display drivers. Probably not this. We did have an even worse form of this same failure mode earlier (where it would crash running literally any game at all) and reinstalling the drivers fixed that, but it hasn't fixed this.
Just to be sure, I disabled Windows Update driver updates, ran DDU in Safe Mode, then reinstalled the main drivers and chipset drivers for this card. They did not resolve or noticeably ameliorate the issue.
Solutions which have been suggested and not tried:
1. Repasting the CPU. I'm not doing this because I did it to the best of my ability last time and I don't have the technical knowledge to evaluate my work and see how it could be improved, so there is no reason to expect better results.
Do you remember peeling off the plastic from the coldplate on the air cooler?
It may be that there's bad contact between the coldplate and the top of the CPU. It might be worth detaching the cooler, repasting and reseating the cooler.
Taking off and putting back the cooler may be enough to fix the overheating. It'd also allow you to look at the paste and see if it covered the whole cpu surface. If not, you mightn't have used enough paste, mounted the cooler improperly or w/o enough pressure, or the cooler could be defective
the very first thing I'm going to recommend is updating the BIOS on your motherboard. it should hopefully have been new enough from the factory to avoid this but early versions of AM5 BIOSes were literally exploding CPUs for the first couple months the platform existed
At least going by the temperature issue, which I've had before, it likely has to do with cooling. If not the CPU paste, it might just be air flow. I've had crashes like this because of too many cables blocking air flow from the fans to rear exhaust in a small case.
Comments
Check seating of CPU power connector (less likely to be this)
Check seating of CPU fan power connector, and ensure that it's in the right header on the motherboard. If it's not in the PWM header, your CPU might not be responding correctly to increased load.
Can also be a RAM issue, try removing one of the DIMM's if it has multiple, or replace with another.
Last, but not least, does the PSU have the ability to give enough juice?
Last, I see someone else mentioned it, how much paste did you apply to your CPU, you might have used too much?
I don't have enough knowledge of computers to evaluate whether the amount of paste on my CPU is the right amount. The amount I put on there is consistent with the assembly instructions I had.
You should not use more paste than the minimal amount necessary to ensure contact over the metal cover of the CPU.
I know many instructions tell you to put like 4 dots, I've always done that, then evenly distributed the paste with a razor blade.
There's also some dedicated CPU and memory test tools you could try such as OCCT.
1.) System RAM is faulty in some way.
2.) The buzzing reeks of a conflict on hardware level; can happen if the GPU is not plugged into the faste PCIe slot (the one next to the CPU is the correct one).
3.) GPU VRAM failure.
Thing is, if it's not the GPU (surprising, given that this problem occurs as soon as videogames are being started up) then I would check the system RAM next. It's unlikely though, because the system runs in the first place.
It doesn't occur on game startup, it occurs after a couple of hours.
You're playing a few games there that are very high performance, the problem is Windows, especially since 11, has so much bloat it's a never ending slog to disable background processes if youre doing anything intensive.
Save what you need externally and install Linux.
ELDEN RING (2022)
CYBERPUNK 2077 (2021)
STAR WARS — JEDI: FALLEN ORDER (2019)
Games with which this hasn't occurred so far include:
THE FORGOTTEN CITY (2021)
THE ELDER SCROLLS V: SKYRIM — Special Edition
THE ELDER SCROLLS ONLINE (2014)
1. Display drivers. Probably not this. We did have an even worse form of this same failure mode earlier (where it would crash running literally any game at all) and reinstalling the drivers fixed that, but it hasn't fixed this.
1. Repasting the CPU. I'm not doing this because I did it to the best of my ability last time and I don't have the technical knowledge to evaluate my work and see how it could be improved, so there is no reason to expect better results.
Do you remember peeling off the plastic from the coldplate on the air cooler?
It may be that there's bad contact between the coldplate and the top of the CPU. It might be worth detaching the cooler, repasting and reseating the cooler.