System stops unexpectedly during high-load testing despite stable temperatures.
System stops unexpectedly during high-load testing despite stable temperatures.
Title remains the same. Consideration for adjustments is needed. Performance stays solid across games and streaming, though it stalls in prime95 and cinebench after a few minutes. No overclocking applied; default BIOS settings are used except XMP. The last crash occurred at 65°C during prime95, while a prior incident hit 50°C. VRM temperatures stayed within limits at 55°C max. Components: Mobo Z690, AMD 12700KF CPU, Gigabyte 3070Ti GPU, Arctic Liquid Freezer 360mm AIO, Corsair Vengeance LP 3600 RAM, RMA 1000x PSU, 4000mm airflow case.
Additionally, there are three fans positioned on top of the case—one at the back, one for the CPU pump fan, two behind the radiator, and three on the front.
It would depend on the location of the issue. This wouldn't be an internal concern. I'm checking if something was left behind and might be causing the problem. For instance, when I swapped my stock ILM with a bending correction frame, I dropped a screw into the CPU area. It was black screw, black motherboard, black case, black cables, and black CPU cooler—nothing visible. I shook it all around and it remained silent. I assumed the screw was removed because there was no rattling noise. After turning on the system, it would play games for about a minute or two before shutting down abruptly. Given the complexity of the CPU power delivery system, this seems reasonable.
Reviewing HWInfo shows your idle voltages compared to stress voltages. A 12v vs 5v difference can lead to unexpected issues if they deviate from expected levels.
Observe that temperatures stayed below 75°C. I've never encountered a PC crashing at those levels. I switched to the balanced power plan instead of high performance to test its effect. It appeared to extend the stress duration by roughly three minutes.