Needing assistance with system freezing and BSOD issues
Needing assistance with system freezing and BSOD issues
I don't see anything helpful in those logs. Notably, Windows didn't shut down correctly... It might be useful to record what the system was doing before the crash on paper. This could help spot any trends. Before jumping into changes, make sure you have important information saved securely. You can monitor temperatures using GPU-z or HWinfo to confirm your cooler is functioning properly. Your power supply unit is also within its recommended capacity for the 3080. Check various voltage readings (through software) to identify any anomalies. Ideally, testing with a more robust PSU would be ideal if possible. (I know Seasonic PSUs are reliable, but reputation alone can't guarantee a fault.) For the next phase, consider reinstalling Windows, as corrupted files could be the issue. And if nothing clarifies the BSODs, try isolating components one by one in another machine—CPU, RAM, GPU, storage. Best of luck.
I'm unable to take screenshots or access external tools like CPU-Z. However, you can open the CPU-Z interface, navigate to the "Motherboard" tab, and capture the desired image yourself. Let me know if you need further assistance!
@Sawa Takahashi Your detailed response is appreciated. So far, I haven’t managed to trigger the bluescreens. No consistent trend in usage or applications has emerged. Occasionally it functions smoothly, other times it crashes repeatedly shortly after. I also performed a 3DMark TimeSpy test and it completed without issues. From my observations, the GPU appears to be throttling under load while the CPU remains stable. Temperatures seem normal, and voltage readings are consistent—12.0V during testing, with minor fluctuations to 11.9. Notably, the "PWR_SRC Voltage" stays around 12.5 and 12.4. I should have mentioned that I’m using an Ubuntu installation as well; freezing occurred there too. No unexpected restarts or blue screens were observed, though usage is infrequent, so it might just be a coincidence. @191x7 Thank you for your effort. Edited August 20, 2024 by Papperlapap spelling
It appears to occur unpredictably, but under what conditions? I experienced random crashes on a 5950X when idle, yet remained stable during heavy loads—until it eventually failed at startup. In the past, turning off software C-States in the BIOS helped temporarily, but that was only a short-term solution. It's true that the PSU can be quite limited for a 3080, though I've used one on an RM650X without problems. Generally, if you overwork the PSU, it tends to shut down rather than trigger a BSOD. Still, the behavior of different PSUs is uncertain, especially when pushed beyond their limits, which might shorten their lifespan.
I expected it, you're running AGESA 1.2.0.6c. It's notorious for instability, USB problems, and fTPM issues. Upgrading the BIOS is strongly advised. AGESA 1207 resolves all these concerns, and the latest versions perform even better.
It seems the data comes from dump files. Memory isn't always RAM, but it's often what people think. Windows moves small RAM chunks to the page file and brings them back when needed, making storage appear like memory. The CPU’s memory controller plays a role; if it fails, it can mimic memory issues. When about half of the dumps point to storage or its drivers, that suggests storage isn’t the main issue here. If there are any overclocking or undervolting problems, remove those components. Updating the BIOS is advisable. You're currently using version F37d, a beta release, which has been removed from the downloads page—likely due to instability. I wouldn't upgrade to F40b; go with F39 instead. To check RAM health, run tests normally with one stick at a time (like during installation). If only one stick causes crashes, the faulty stick is the culprit. If it fails with either stick, the CPU might be the problem. Memory testers often miss defective RAM, especially DDR4 and newer types.
At the beginning, thank you all for your support—it really means a lot! Second, I suspect we're encountering a new issue: during roughly 1 in 7 restarts, the system fails to launch far enough to reach the boot screen. It just stays black and the CPU fan doesn't activate as it should. @Alex Atkin UK This occurs no matter what I do. Or at least that's what I'm observing. It freezes, crashes, or triggers a BSOD when I respond here or open a text file. Sometimes it works fine, enduring even a full 3D Mark Timespy test. Or Baldursgate 3 character creator, hard to say which is more challenging. Also, checking power consumption shows my PC and two screens rarely exceed 500W. In fact, they usually stay under 400W, even during gaming. I cap my FPS on the NVIDIA Control Panel at 65 FPS. I turned it off when the crashes began, hoping to reduce power use. @191x7 I'll update my BIOS as you advised right after posting here. Thanks for the tip! @Bjoolz I haven't overclocked or undervolted anything. I plan to update the BIOS and test your RAM theory. I'm not sure what you meant by Windows reinstalling, since freezing happened on Ubuntu too. So I assume Windows isn't the cause here. Please confirm if that's correct, it's a tough question.
The issue remains unclear to me about triggering the crashes. I'll check the benchmark again. I'm planning to share updates tomorrow or later if anything occurs. Just completed the BIOS update and will begin with RAM testing. Appreciate your advice, enjoy your rest of the day!