Game crash occurred without BSOD or error alerts
Game crash occurred without BSOD or error alerts
Hey, I’m trying to figure this out. I recently built a gaming PC and sometimes during intense gameplay I see a BSOD (black screen). The display goes dark, audio cuts off instantly, then the system briefly loses power before coming back on—usually with the motherboard LED or VGA light flickering on, and I have to press the power button to restart. There’s no error code shown in Event Viewer (Kernel-power ID 41). It mostly happens while playing demanding games like HLDS but not in League of Legends.
Details:
- OS: Windows 11 64-bit (also worked on Windows 10)
- GPU: AMD 5600X
- Motherboard: MSI B550-A Pro
- RAM: Crucial Ballistix DDR4 3600MHz, 2x16GB
- PSU: Enermax Revolution DF 750W Gold
- Specs: Latest drivers installed (version 7C56vAF)
I’ve checked the GPU placement multiple times—seems correct—and changed the power cable setup from one to two for the dual GPUs, but it didn’t affect the crashes. I also ran GPU-Z and a stress test with AMD Adrenalin, but no crashes occurred. The logs are attached if you want more insight.
Anyone else notice similar issues? Any clues could help me diagnose this.
Hello, thanks for your input. From what I recall, the crash occurred at both 3600Mhz and 2667Mhz, likely before enabling XMP. I might verify that. Also, the HWinfo log from Thursday is available as a CSV file.
Join the discussions here! A sudden shutdown without BSOD suggests a power problem. Probably the PSU is failing.
Update briefly, I disabled XMP and managed to play for about half an hour without any issues. Before, crashes usually happened after around 10 to 15 minutes. I'll keep watching to check for any changes. Thanks for the introduction and support. I also considered the PSU might be the problem, but I recall these issues started early. I bought the PSU new, so maybe it was the setup—using just one cable for the 2x8 pins or possibly linked to the GPU.
I'll check that out. After more testing, the problem still exists. It appears less common now, which is good... Would running tests like memtest86 or cinebench assist in identifying the issue? Also, as a final option, I have a backup power source I can switch on and see if it resolves the matter (it's just frustrating having to disconnect/plug everything).
You might want to use Passmark Memtest86 or Memtest86+ to verify your RAM. OCCT, Prime95 or Cinebench can help determine if CPU stress triggers the crash. Furmark, Unigine Heaven, MSI Kombustor will assess GPU load issues. Combining both CPU and GPU stress tests offers a comprehensive check. OCCT also includes a PSU test for this purpose. For drive diagnostics, Hard Disk Sentinel is a useful tool.
Hello once more, 191x7, your advice was really helpful. I finally dedicated some time to stress-testing: Memtest86+ ran slowly but passed without issues; Cinebench24 multi-core stayed stable up to about 110W; Furmark hit a peak of 270W without any temperature problems or crashes (lasted roughly ten minutes); Cinebench+ took around ten minutes before crashing. I’ll check the PSU test from OCCT and consider swapping it if needed. Thanks again! Updated: Completed OCCT power test for an hour—CPU reached 125W at 95°C, GPU used 230W at 70°C, no errors found.
The processor runs much hotter than expected for a 5600X. What cooling solution should I use? Checked the Hard Disk Sentinel.