Procesor power depends on the kernel's efficiency and hardware capabilities.
Procesor power depends on the kernel's efficiency and hardware capabilities.
My computer has experienced random freezes that prevent it from recovering, starting from 2022. It tends to crash more frequently during demanding processor tasks after about 30 to 90 minutes. It seems to require a "cooldown" period before I can restart it and resume work, as it restarts again within minutes. Temperatures aren't the problem since CPU, GPU, and VRM are all within normal limits. The Event Viewer doesn’t provide useful details either; it only shows the critical error "Kernel-Power-41". "Kernel-processor-power" appears only as a warning. My system specs include an I5-12600KF Deepcool GAMMAXX L240 MSI Tomahawk B760 DDR5 (previously Gigabyte AERO Z690 DDR5), Kingston FURY Beast 16 GB (2 x 8 GB) DDR5-5200 WDC PC SN530 SDBPNZ-256G, Asus STRIX GAMING OC GeForce RTX 2070 SUPER 8 GB Corsair RM750e (previously RM750x). I initially thought the issue might be with the motherboard or power supply, but replacing them didn’t help. I’ve also tried adding RAM sticks and reinstalling both Windows 10 and 11 multiple times. The boot drive shows no problems in CrystalDiskInfo, and both BIOS updates and resets have been completed. I’m really puzzled about what’s causing this, and I appreciate any assistance.
I've never encountered a KP41 event dropping below Critical. When everything freezes, the behavior is puzzling. Fans still operate? Display remains active? Keyboard and mouse lights still glow? Sound plays? Or does it simply shut down completely? The repeated crashes suggest a thermal problem. I’d suggested cleaning the CPU, but you’ve already replaced the motherboards—did you also install fresh thermal paste? Where are your temperature readings coming from? Are you using HWMonitor or Info? And do you monitor them while attempting to force a crash? Also check the AIO’s RPM. There’s a possibility it’s failing internally. I’d run Prime95 or Cinebench in a loop with continuous logging via HWInfo to observe trends. Keep an eye on CPU temps and pump speeds to ensure stability and prevent fluctuations. There’s a risk the 2070 is struggling, possibly due to aging hardware rather than GPU load.
Initially, thank you for your reply! When the computer "shuts down," everything remains active, the display freezes, fans light up and spin, while the mouse and keyboard stay responsive. The sound stops. You noted my GPU needs a refurbishment, which I’ve already done, along with replacing the thermal paste on the CPU after switching OSes. I usually track temperatures using the Nvidia overlay, but I also used the MSI Afterburner overlay before. (No overclocking is involved; I rely on Afterburner for custom fan profiles.) Since the GPU is limiting the CPU performance (I’m running at 1440p), I rarely see temps exceed 45°C, with the CPU staying under 50% load—usually around 27°C. For the GPU, it’s typically 72°C during heavy use and drops to 39°C when idle. You provided a useful suggestion: the crashes seem worse in more demanding games like HOI4 and CS:GO compared to CS2, which is more GPU-heavy. I could monitor temperatures during Cinebench to rule out AIO issues. Would you think it’s likely the CPU has a problem?
Intel didn't fail with the 12th generation as they did with the 13th and 14th models. I think the issue lies more with the graphics card than the CPU. My prediction is that the graphics card is the problem. Considering you've already updated the motherboard and power supply, do you have a UPS available?
No, I don't have an UPS. I've already attempted to run Cinebench on both multi- and single-core settings, and the temperatures stayed well below 58°C. No crashes or freezes occurred. I could use a GPU temporarily, but it might take a few days. In the meantime, I can check another power source. I suspect the problem isn't the outlet itself, since the crashes aren't consistent—especially not during demanding tasks like War Thunder.