Hey everyone, I've been dealing with a strange issue regarding my PC since last year. It tends to crash suddenly under load, and while the event viewer mentions potential driver or RAM issues, it often just goes blank without rebooting. The fans and RGB lights stay on, but the RGB defaults to a basic color instead of what I set. After running stress tests with FurMark, I've managed to get it to run for about 20 minutes before crashing again. One day, I thought it was stable but returned from dinner to a headache-inducing burnt plastic smell. I checked internally and found that the GPU smelled strongly of burnt plastic. I sent it back under warranty, but after a week, they reported it was fine. Now that I'm ready to troubleshoot again, I tried FurMark and noticed the smell was fainter, but the crashes returned along with new event viewer errors regarding nvlddmkm. I'm out of ideas here; could it be something like thermal paste on the CPU that needs to be changed?
2 Answers
That burnt plastic smell is a big red flag for faulty hardware. Since you're seeing issues with nvlddmkm.sys, it points towards the GPU. Make sure the power cable is securely plugged in and check for any visible damage like bloated capacitors. If you can, try replacing the GPU to see if that solves your problem. If you're hesitant to buy a new one, consider taking your entire rig to a repair shop for diagnostics; they might pinpoint the real issue without a full fix. You could then use that information to push for repairs under warranty if the GPU is indeed at fault.
It sounds like the power cable to your GPU might be the culprit. High-powered GPUs can end up drawing so much power that the cable itself melts or burns. Definitely check there first to see if that's the problem!
Related Questions
Lenovo Thinkpad Stuck In Update Loop Install FilterDriverU2_Reload