All three displays turn off and the computer stops working during games (black screen of death).
All three displays turn off and the computer stops working during games (black screen of death).
OS - Windows 11, roughly two years old. Installation process also took about two years. CPU: AMD Ryzen 9 5950x, GPU: RTX 3090, Motherboard: PRIME X570-PRO, PS: EVGA 1000W EK 295. Vanquish Desktop PC is a custom watercooled build. I attempted cleaning the PC, re-seating the GPU, and performing a driver clean install, but the problem persisted. Currently, when playing games like POE 2, my monitors suddenly go dark. Initially, it happened without me playing—just after finishing a session. I still hear some audio briefly after the screen goes black, but eventually it disappears completely, requiring a forced reset. I attached perfmon and diagnostics files with minidump data as per the thread instructions for BSOD. perfmon (2).zip
Welcome! The ongoing noise will simply indicate a buffer emptying event—everything on the PC will crash simultaneously as the screens go dark. If your temperatures are normal, it likely points to a power supply issue. Without testing it directly, it's hard to diagnose, which is a significant task. It would be wise to reinsert your RAM and CPU to rule out a faulty connection. If that doesn't resolve the problem, I'd focus on inspecting the PSU next.
BTW, BSOD stands for Blue Screen Of Death, often linked to kernel or driver problems, though these can stem from hardware faults. It’s a "graceful" shutdown, allowing data collection. A hard crash usually lacks logs, making it harder to diagnose. If a recent change—such as a driver update—correlates with the issue, reverting might help. Consider using System Restore points to check if they precede the problem. However, since it’s an ungraceful exit, I’m skeptical about a software cause.
The PSU seems unlikely to fail after two years—it should stop working right away or after many years, not in just two years. If it does, reseating components might be simpler than trying to revert drivers. Still, the RAM is worth checking too. Repeated heat expansion can sometimes damage contact points, and a new installation can fix the issue immediately.
I'll check the PC and reinsert the RAM. It seems the CPU won't work because I need to empty the loop and there isn’t enough fluid. Could a hard drive problem be the cause? I’ve seen the crashes happen only when using the hard drive instead of the SSD, so more testing is needed to confirm.
If it were an HDD, I’d anticipate a freeze instead of a complete crash. The crashes seem to occur during gameplay, not at startup. I’d only imagine a hard failure if the drive was spinning too fast and overloaded the power supply—something that’s improbable on its own.
Updated drivers were applied, Windows reinstalled, and RAM re-seated. This seems to have helped a bit but the issue persists. My screen flickers black several times before returning to normal, almost as if nothing happened. The reliability monitor shows a specific error code. I suspect the livekernal 117 component might be involved; it was reseated already, yet the problem remains. Any suggestions?
Well, now you're receiving updates! It's strange to see improvements in this manner yet still not resolving everything completely. What an odd situation. Have you attempted using just one monitor or a different, possibly branded, second monitor? If only one is connected, does the type of monitor matter? In short, I'd experiment with various methods to capture the output—check GPU ports separately, use another PCIe slot if available, test on another machine. I'm acknowledging it's less likely the power supply is the issue, but it doesn't seem related to Windows or drivers now. Was everything done at once or were you testing after each change? Was the Windows reinstall the cause, or did the RAM installation affect things? If the latter, consider running Memtest86 with at least two cycles.
I completed all the mentioned steps at once, which might have significantly enhanced the performance. Based on my experience, the most probable fix was reverting the GPU driver since the other one wasn’t stable and I’ve faced driver issues before. Now I plan to run Memtest 86 on the RAM.