Is it happening that my computer stops unexpectedly during gaming sessions?
Is it happening that my computer stops unexpectedly during gaming sessions?
my pc has been shutting down unexpectedly during gaming sessions. Around ten minutes into the game, it would sometimes restart or stop completely. I checked the event viewer and found a critical error labeled "kernel power" with event ID 41. A window minidump is available for those using a debugger. The system's BIOS is up to date, and the GPU drivers are current. Memory tests and Cpuid results show no issues with RAM. I recently replaced the CPU, RAM, and GPU, rearranged the cables, and verified everything fits properly. The GPU temperature stays between 50-70°C during heavy use (25-30°C when idle), while the CPU stays around 45-60°C under load (also 25-30°C idle). My specs include Windows 11, Ryzen 7 2700x, an AIO 420 Aurora 3060 GPU, B350 Tomahawk motherboard, 32GB DDR4 RAM at 2133MHz (4x8GB), RTX 3060 graphics, and a 600W PSU with EVGA 80+ certification. I suspect the power supply unit might not be delivering sufficient power for the new GPU, though the debug light on the motherboard points to the CPU as the issue. I also noticed some crashing problems with my old RX580 even before installing the 3060, which was installed around Christmas. The 3060 itself appears to be functioning correctly. Any suggestions? Thank you ahead of time.
Can you ensure the dump is set to public in Google Drive? (Confirm 'anyone with the link').
I think this might be the reason for the issue...
Check the temperature monitor (such as HWMonitor or CoreTemp) and observe the CPU temperatures. It’s possible it shut down because of overheating if the repaste wasn’t done properly.
it's now public, thanks for keeping me updated.
i'll start running hwmonitor/coretemp just to confirm.
edit : hwmonitor/speccy screenshot attached
everything looks good. after about 5 minutes of benchmarking.
edit 2 : crashes occur every 4th or 5th try. temperatures are normal, cpu stays between 45-50 throughout each test. except the gpu which reached 68-70 – that's the ideal range for it under load?
This PSU isn't suitable for these specifications—it's a low-quality, group-regulated unit with poor build. If it were brand new, it wouldn't fit either the GPU or the newer models. Using it now would only make things worse. Removing it is essential to accurately troubleshoot the problem; it should have been replaced even without crashes.
I believe the CPU is extremely powerful and it's putting a lot of demand on the VRM from your motherboard. The B350 Tomahawk has that setup, but the board tends to become very unstable and slow.
I rolled back to windows 10 during a clean reinstall because I detected corrupt system files.
Even after the clean install, the same crashing problems persist.
I disabled automatic restart so I could perform a BSOD to capture a dump file, but it doesn’t work since the PC shuts down without it.
I’m now uncertain about the source of the issue.
I suspect the PSU might not be supplying power when the GPU is needed.
However, I noticed that the motherboard’s debug lights turn on for the CPU during crashes, so it’s unclear who is the main culprit...
I’ll update with a new hardware monitor showing temperatures in a few minutes.
(Just woke up and need coffee first 😉)
Do you believe switching the PSU would fix this situation for you? I’ve been considering the Corsair RM750x (2021). I think a PSU would be enough to handle everything I can use, plus a bit of extra capacity? Yeah, it’s 2700x... I made a typo there. I managed to run the B350 with a 2700x for several years...around the fifth year without any problems, but why now?