Issues Arise in OCCT Power Testing Procedures
Issues Arise in OCCT Power Testing Procedures
Hello everyone, I'm encountering a particularly stubborn issue with my PC, leading to crashes under specific conditions, and after several diagnostic attempts, I'm seeking insights from this knowledgeable community. y system consistently crashes to a blue screen showing the Stop Code: CLOCK WATCHDOG TIMEOUT, particularly during the OCCT power test. This issue does not occur during other OCCT stress tests, such as Linpack, CPU, and GPU tests. The crashes also happen when engaging in power-intensive tasks, such as running AI models or video processing with ffmpeg, suggesting an issue with power management or thermal performance. For further analysis, here are the dump files I've collected: https://files.catbox.moe/hmet8g.zip and the windows report htmls: http:// https://files.catbox.moe/jfoubg.zip System Specifications: CPU: Intel i7 12700KF GPU: RTX 3080 Ti 12GB RAM: Corsair Vengeance 64GB (2x32GB) DDR5 5200Mhz PSU: Corsair RM850X 850W Gold Cooler: Lian Li Galahad, currently not performing optimally due to a manufacturing defect Troubleshooting Steps Taken: Memory Testing: I've individually tested each memory stick in various slots, which did not resolve the crash issue. Cooling System: I'm aware that my cooling system isn't functioning properly. I'm currently waiting for a replacement cooler from the manufacturer. However, given the crashes occur specifically during the OCCT power test and not under temperature stress tests, I'm inclined to believe that the issue extends beyond just cooling. BIOS and Drivers: All system drivers and the BIOS are fully updated, with settings reset to defaults. My system is not overclocked. Power Supply Considerations: I'm pondering a PSU-related issue and am considering using a PSU tester for a deeper diagnosis. I also have a spare Seasonic PSU (serial number "RT9026Z00R6ZGBB0125") for testing, although I haven't identified its model yet and am cautious about disassembling another PC without being sure it's the right move. Peripheral and Power Configuration: I've also tested the system with all USB devices unplugged and ensured it's plugged directly into a wall outlet, eliminating any variables that external devices or power strips might introduce. Given these circumstances, I'm puzzled by the direct link between the OCCT power test-specific crashes and my system's configuration or hardware. The specificity of the test that causes crashes, despite high temperatures and cooling issues, leads me to suspect there might be underlying issues related to how my system manages or is supplied power under load. Before I proceed with potential PSU testing or swapping and before the arrival of the replacement cooler, I'd like to gather any advice or insights you may have. Could there be an overlooked aspect or a component sensitivity to power variations not evident in other tests? Thank you all in advance for your time and help. Any suggestions or insights would be greatly appreciated!
Focus on improving the power supply unit and ensure at least 1000W capacity. Consider using Trident Z/ Samsung B-dies RAM instead of the Vengeance option. Note: You might want to test your RAM at standard speeds first, without any overclocking, just to see if it makes a difference. This seems likely to be related to either the PSU or the RAM issue.
Confirm if the newest BIOS and chipset drivers are up to date.
Hey, Thanks a lot for taking the time to respond. I've actually already run the RAM at stock speeds since XMP is disabled in my BIOS settings, ensuring they're not overclocked in any way. Also, I've tested each RAM stick individually and in different slots to try and isolate if one might be causing the issue, but unfortunately, that hasn't shed any light on the problem either. I understand your point, and it makes sense that the issue could be tied to either the PSU or RAM (or both). I'm just hoping to find a more definitive way to diagnose the issue before proceeding to replace parts. If you have any thoughts on how to further test or isolate the issue with the tools or methods I might not have tried yet, I'd really appreciate hearing them. Thanks again for your advice!
Execute a concurrent GPU and CPU stress test; if the system fails, indicate a PSU problem.
A PSU problem becomes clearer through comprehensive system testing rather than isolated CPU stress tests. A failure would likely trigger a full shutdown to safeguard the device, which wouldn’t display a bluescreen. If the system restarts with a blue screen, other causes are more probable, though a PSU issue is still less likely. The description mentions a "power intensive task" but screenshots show minimal GPU usage and only about 200W CPU load—just a quarter of the PSU’s capacity. I suspect this isn’t PSU-related. I’ve reviewed all five minidumps; four showed watchdog timeouts and one had an unhandled exception, pointing to software/driver problems. The KMODE error points to "FocusriteUsb.sys" and possibly a DAC or audio driver conflict. Uninstalling/reinstalling drivers might resolve it, or interference from multiple audio drivers could be the cause. If crashes are confined to OCCT, don’t stress too much—be alert for issues outside that environment.
Four crashes involved the CPU freezing and one was due to Focusrite. The Focusrite seems to trigger BSODs often, so reinstall its driver. These issues appear unrelated. I’m not very familiar with Intel boards this year, so it might help to verify if they offer built-in overclocking similar to AMD’s Precision Boost Overdrive and turn it off. The last possible cause could be a defective CPU. This isn’t the typical crash from a bad CPU, but it’s not rare either. If it happens during a power test, the problem might lie with the motherboard’s VRM.
Hello everyone, thank you very much for taking the time to join me through this discussion. Concerning the simultaneous GPU and CPU stress tests, it seems that the OCCT power test, which triggers an immediate shutdown or restart, puts pressure on both components at once. This aligns with my observation that the problem could point toward a power supply issue. Nonetheless, our conversations have sparked some thoughts about whether this is definitively a PSU concern or another factor at play.
About the PSU possibility: I’d like to clarify a few points regarding the screenshots and power usage data. The images I provided are solely from the OCCT tests—covering Linpack, GPU, and CPU stress scenarios—not from when I’m running demanding tasks like AI processing or video editing with ffmpeg. The power test in OCCT, where crashes occur instantly, doesn’t last long enough for me to collect any meaningful metrics, as my computer freezes right after launching.
Regarding the Focusrite DAC driver (FocusriteUsb.sys): It’s worth noting that this file shows up in the minidump logs shortly after the PC rebooted during OCCT. At that moment, the system is still loading initial drivers, including FocusriteUsb.sys, which can take time. This delay might be why it appears in the diagnostics rather than being the root cause of the BSODs. To confirm, I plan to uninstall and reinstall the driver, then rerun tests with the device disconnected. This helps rule out other variables instead of addressing a confirmed problem.
When it comes to potential software or CPU concerns: checking for overclocking features and inspecting the CPU or VRM on the motherboard is sensible. I’ve verified that no overclocking is active, and both CPU and motherboard settings are set conservatively for stability. While the system handles other stress tests without immediate failure, the sudden crash during OCCT pushes me to dig deeper into what might be behind such a sharp reaction. It’s also important to mention that similar crashes occur when performing power-heavy tasks outside of OCCT, like using Retrieval-based Voice Conversion or running ffmpeg. The problem isn’t limited to OCCT tests alone.
Thanks again for your support. I’ll continue working through these steps and will keep you posted on any updates or new insights. Feel free to share any additional ideas or observations based on this information!
I wanted to share an update and appreciate your suggestion to use a PSU tester for troubleshooting my PC. I purchased one and verified the 24-pin motherboard power, CPU cables, and GPU PCIe connections. All tests came back normal. Notably, 3D rendering tests showed smooth performance on the GPU but a sudden fan speed cap and blue screen when switching to CPU mode. This strongly points to a cooling issue, especially considering known problems with Galahad AIO units that LIAN LI has mentioned. I’m now exploring replacing the AIO, as it seems to be the root cause. It’s been quite an experience. Thanks again!
Hey, I understand your concern. Have you checked if the problem might stem from insufficient cooling? Your setup isn’t top-notch, but some research suggests it could be due to a motherboard issue—like a bad VRM—or possibly a CPU problem. Any updates would help! Thanks!