Help! New PC randomly shutting down
Help! New PC randomly shutting down
Hey there, this is my initial build in what feels like a long time and I'm trying to figure out why my PC shuts down under heavy stress. I've got a Z690 motherboard with EDGE Wi-Fi, DDR4 RAM, a 13th Gen Intel i7-13700K processor, 3400 MHz, 16 cores, and a GTX3060 GPU. The power supply is RM850x 2021, and I'm using a Noctua NH-D15 cooler with two fans. I've been running Furmark stress tests on both the CPU and GPU to see how they handle load. Everything works fine at default settings, restarting in about 2-3 minutes.
Do you think I might have made a mistake while assembling? Like putting the CPU and GPU on separate rails? Or could it be that the VRM isn't delivering enough power? During the tests, the CPU was drawing over 250 watts, which is quite high. I also limited the power draw to 200 watts using throttle stop, and the system stayed stable for the past 30 minutes without crashing.
Did you check if your motherboard's BIOS is up to date? I noticed you entered 7D31—could you confirm which release it corresponds to? The latest stable is 7D31v1A, which includes the newest CPU microcode updates. It might help if you’re not already current.
I just made the bios update to the newest release and it seems more reliable—it handles the same furmark CPU burner and GPU stress test for at least 30 minutes without issues. Still, I notice occasional crashes during regular use, especially when mining overnight while the computer is idle. The core temperatures are consistently around 90°C according to coretemp. Could you share a guide or links on using Event Viewer properly? I attempted it myself and a quick Google search only led to more confusion.
The 90-degree setting feels warm, but it should stay within acceptable heat limits. I understand CPUs tend to get hot. In the Event Viewer, go to "Windows Logs," then select "System." Look for "Filter current log" on the right side. Check the "Warning," "Critical," and "Error" levels and click OK. Are there any entries showing messages during crashes? If yes, record the Event ID numbers and share them here. This should help identify the issue. It’s also a good idea to run memtest86 (MemTest86 official site) to verify RAM health and rule it out.
Furmark has often been labeled a power virus, which is understandable given its behavior. It can trigger shutdowns or reboots due to the strain it puts on systems, especially when used for burn-in testing. For stress testing purposes, I recommend alternatives instead.
Reviewed the event log immediately after the crash and saved it here. It doesn’t appear linked to a software failure. I’ve also tested RAM without XMP settings, both individually and in pairs, which causes crashes within a minute. I’ll run memtest and share the findings. The two stable setups are: 1) stressing CPU and GPU separately doesn’t trigger crashes; 2) gently reducing CPU voltage by 0.1V and lowering clock speed to 3.6 GHz also prevents crashes. You might want to consider power supply or motherboard problems, possibly due to a connection issue. Any additional tests could help narrow down the cause? Thank you for your patience!
If memtest didn't reveal any issues, I believe the problem likely lies with the motherboard or power supply unit. Verify that all motherboard power connectors are securely attached and functioning with the PSU. Initially, try replacing the PSU, and if it resolves the issue, proceed to replace the motherboard. Check Event Viewer for any related errors chronologically.