Reduced crash rate with 1% GPU adjustment
Reduced crash rate with 1% GPU adjustment
Hello, I’m trying to figure this out. Yes, Windows 10 is no longer supported. It’s been acting like that for over a year now. I’ve tested everything I could find and changed all the hardware. My friend is facing the same problem, and I’m really stuck with no clear solution. Occasionally the system slows down gradually. Programs stop responding as if they’re not working at all. There’s no sudden crash; it just becomes unresponsive over time—usually 1 to 3 minutes before I can’t restart and have to turn off the power. Sometimes it takes weeks between incidents, other times it happens multiple times in an hour. I haven’t identified a common cause yet. Event Viewer shows no repeated errors or critical issues. The temperatures look normal: CPU around 41°C, GPU 52°C, all RAM temps between 30–48°C according to HWiNFO64. It seems the graphics card might be involved because Task Manager displays GPU usage as -1%, which is negative. This doesn’t happen when the GPU is active—like during games or heavy tasks. I haven’t found a problem with the GPU itself or its drivers, since I’ve tried multiple cards. I’ve shared this before on Linux forums – it was about a riser issue ( I’ve been trying different setups recently: - SSDs: Samsung 970 (2020) and Intel 660P (2022), both fine. - GPUs: RX 6900X, 1080Ti, RX 6750X. - CPUs: 5800X, 5600X. - RAM: 16GB G.Skill Neo 3200, 32GB G.Skill Neo 3600, 32GB T-force Xtreme (2 or 4 sticks). - Motherboards: Two models, with and without riser, PCI-E Gen 3/4, multiple slots. - Power supplies: Corsair RX1000 and Silverstone 850W. - Drivers: Multiple versions of Windows 10/11 reinstalled; removed apps like Teams, PhoneLink, Logitech G-HUB, etc. - BIOS tweaks: XMP/DOCP enabled/disabled, downclocking, secureboot, etc. - Clean boot attempts and stress tests (Furmark, 3DMark, OCCT). - Driver updates and disabling MPO. I’ve also tried changing RAM sticks, cables, and even swapped out the CPU and GPU. Still nothing consistent. If anyone knows what might be causing this or has similar experiences, it would be really helpful. Thanks a lot!
Indeed - well, I've tried various combos over the years. The only thing that's the same since 2021 is the case... Even the fans has changed; Noctua, Corsair, Lian-Li -with different controllers. But for the life of me, how could that be the case? It's not shorting with that kind of crash?! All of the pieces except for GPU has been in there since first time it did it, except the RAM, GPU and fans.
It seems likely the issue involves Windows or software, as all hardware has been checked. You're considering extending the current build for another 2-3 years, but you're thinking about creating a completely new PC and then booting from the old SSD with Windows installed—though this could be costly.