Occasional power cuts using adjusted processor currents.
Occasional power cuts using adjusted processor currents.
CPU model: Ryzen 5600x AIO, cooler: MAG CORELIQUID 360R, GPU: RTX 3070, board: ROG STRIX B450-F GAMING (newest BIOS 4602), RAM: 8GB. Used Skill RipJaws DDR4-3200 DIMM, 16 slots, PSU: EVGA SuperNOVA 650 G2, 80+ GOLD 650W. System specs: W10 21H2 19044.1415. I've had this setup for about a year. I adjusted manual voltages, turned off PBO, and set the CPU to 4.3GHz because of excessive voltage from the motherboard when PBO was on (4.5-4.6GHz, sometimes up to 1.4V+). This caused temperatures to spike to 65-70°C even at idle. After stress tests, temps dropped to 40°C at idle and around 70°C during Cinebench, with no shutdowns. Initially, I started at 1.1V and 4.3GHz; everything worked fine. The next day, random PC crashes occurred even without CPU load, but temps stayed normal (~40°C). I increased the voltage to 1.2V and kept 4.3GHz for a few hours, which resolved the issue temporarily. Eventually, I settled at 3.7GHz with idle temps at 33°C and 46°C during stress tests, still no shutdowns. I’m trying to figure out how to optimize this without sacrificing performance or risking overheating. Older BIOS versions seemed more stable. What’s the best way to find safe voltages that keep everything running smoothly? And why do things work fine when temps are normal but fail under stress? Is changing only the motherboard enough?
Previous models managed voltages more effectively... I doubt 70°C with PBO turned on at idle while using an AIO is typical. It seems odd since it doesn’t crash during stress tests, only when CPU usage hits around 5-10%, making it tough to get the right voltage and causing occasional shutdowns afterward.
It could be that PSU experienced another unexpected power cut due to the stock configuration.