PC keeps crashing
PC keeps crashing
My system was assembled in 2019 and functioned smoothly until last year when my software started failing repeatedly (screen went dark, had to force shut). I attempted updates, changed drivers to various versions, rewrapped the graphics card, but nothing resolved the problem. Once I switched from Windows 10 to 11, the latest driver updates seemed to stabilize the crashes. Each time I restart, the AMD utility alerts me that drivers crashed and reset settings, though it doesn’t actually crash now. During normal use it runs without issues, but demanding games cause sudden shutdowns—no black screens this time, just instant failure. Checking Event Viewer revealed errors linked to the game input service, especially after Windows 11 installation. After removing the older driver version, the issue persisted, but the Event Viewer now remains silent. Could the power supply be faulty? It should provide sufficient power, and I even reduced the GPU clock speed. With multiple recent overheating events, might there be a hardware concern? The GPU model is 5700 XT, temperatures typically range between 60–80°C, which seems acceptable. CPU temps are lower but still high at 3600 RPM—around 70°C idle and up to 85°C under load. I recently replaced the case fans; they cut temperature by 10–15°C when the side panel is off. I apologize for the long explanation, but with limited details, I’m trying to describe the situation clearly. Here’s my PC: https://pcpartpicker.com/list/9m7Dfd
It seems to be a RAM, GPU, power supply, or software problem. The simplest step is to test one RAM stick at a time—install it, run a stress test, then swap it out. Next, try installing Windows and drivers on another drive or a flash drive for a quick check. For the power supply, swap it with another unit if possible. GPU hardware follows the same process. Check your temperatures; see if other components are struggling. Overclocking isn’t recommended unless you’re sure. Resetting the BIOS can sometimes resolve issues. My Intel Iris XE had strange behavior where a driver crash caused screen distortions until a cold boot fixed it—likely a BIOS or driver conflict.
I attempted to remove the side panel and experienced immediate failure. XMP was disabled, then it crashed again instantly. Gradually removing one stick at a time helped for a while, but the 4th stick caused total failure—drivers had to be uninstalled via Device Manager. After uninstalling, testing with drivers removed still led to crashes with every stick. The issue occurred randomly, not consistently tied to a specific stick. After updating drivers, it kept crashing instantly when loading into the game, regardless of stick. I restarted the PC over 20 times without finding a clear pattern. It seems RAM could be involved, though I couldn’t isolate a single stick as the problem source. The timing aligns with RAM changes around 1.5 years ago, but it also failed with older sticks, making it hard to pinpoint. My tests aren’t definitive, but the consistent, sudden crashes point toward RAM rather than hardware failure.
The RAM tests seemed more like a temporary solution. Removing the GPU and switching to another one or using integrated graphics might help. A driver issue will keep crashing even after an OS update if the hardware stops responding to the driver. Based on what you've done so far, the GPU is likely the problem.
Lack of a spare GPU is making testing difficult. The crashes and freezes have intensified recently. Now even partial loading triggers failures—screen freezes without full load, forcing shutdowns. It also crashes when starting up, sometimes showing no signal at all. After a few forced restarts, it restarts normally. Audio stutters before the crash appears. Likely GPU issue. Undervolting didn’t help. I might replace it and return it if it doesn’t work. Another odd pattern: crashes instantly under load, but sometimes never—until restarting, then it resumes. This inconsistency matches what I see during RAM tests, but I’m unsure what it means.