Is there another issue with the kernel power 41 troubleshooting?
Is there another issue with the kernel power 41 troubleshooting?
Hello everyone!
I’m sure you’ve encountered this exact situation before, but I’m sharing my details here to help troubleshoot more effectively.
Problem: In games, the computer suddenly blacks the screen and then restarts almost immediately. This has occurred intermittently (every few days) since I upgraded my PC.
System specifications (after upgrade)
B550 Aurora Elite (rev 1.0)
Ryzen 7 5700X3D
Thermalright Phantom Spirit 120mm EVO
32GB (4x8GB) 3200MHz RAM (g skill ripjaws)
SK Hynix 500GB SSD
WD Blue 1TB HDD (5400rpm, according to me)
RTX 3080 (this was a certified refurbish card)
PNY XLR8 10GB
Cooler Master 750W Gold
2x Lian Li 140mm fans
3x Lian Li 120mm fans (both sets are standard ARGB, not the fan mods)
From what I understand, I’m not hitting the power limit. Under full load, GPU is around 320W and CPU about 50W.
My main concern is that unless I push the system to extreme overclocking, I can’t reproduce this issue during stress tests. Even temperatures are normal—GPU at ~72°C, CPU around 68°C under full load in 3D Mark.
I have an overclock enabled in Afterburner and PBO is active, but it doesn’t seem to affect the shutdown frequency.
It typically occurs after a long gaming session, maybe a few hours later, but again, temperatures remain stable.
I might try testing the PSU voltage, though I don’t have a multimeter and nothing unusual shows up in Hardware Monitor.
My current plan is to run memtest tonight. I’ll post the results here. I’ll also reset BIOS settings (current version), clear MSI Afterburner, run DDU, and turn off PBO. I’ll double-check the cable connections and latches on the GPU and motherboard.
Please let me know if you have any advice or additional tests to try. I can share minidump files or HWM values if needed. I’ve heard that using 4-channel RAM might benefit from disabling XMP and adjusting overvoltage to +0.01V.
Thanks, everyone! I appreciate your help!
I would blame the PSU, the 3080 has extremely bad power spikes that cause it to fail.
These spikes last only milliseconds, so software doesn’t detect them. The 3080 is rated for 320 watts, but the spikes can double that, tripping the 12v rail like hitting the reset button on a computer.
If you keep using an older ATX2.0 PSU, go with a 1000w. For the newer ATX3.0, 750w works fine; 850w would be ideal. The ATX3.0 comes with updated specs that manage GPU spikes much better.
UPDATE 1: so I ran memtest and got zero errors, and I let it run every pass. So, my two options now are essentially the GPU and the PSU. The disks seem fine, and under stress testing the GPU is fine. It will do 3dmark and Furmark with a baseline overclock without even breaking 72 degrees under 100 percent usage and 345W. So at this point I 100% agree that is most likely transient spikes from the 3080. As of now the plan is to grab an ATX3.0 PSU, I'm thinking the Thermaltake or MSI 1000w gold. They are not that much more expensive than an 850w and gives me some more headroom.