Question about GPU drivers causing immediate system crashes, with multiple GPUs tested.
Question about GPU drivers causing immediate system crashes, with multiple GPUs tested.
Hello,
I’m seeking some guidance regarding a peculiar problem I’ve encountered with my system. During a game last week, my computer suddenly crashed. Over the weekend, the issue intensified, causing it to fail whenever any GPU driver was updated. Initially, I suspected a faulty GPU (my old 970 model), but after replacing it with a new AMD 9700 XTX, the same problem persisted. The system operates normally until I activate a non-Microsoft Basic Display Driver, at which point it crashes immediately. Restarting via safe mode or using a system restore resolves the issue. Device Viewer sometimes displays the PCIE device correctly but always indicates something is present in the port. It seems there might be a faulty motherboard or incomplete driver installation following a fresh Windows install.
I’ve tried several solutions without success:
- Fresh installs of Windows 10/11 twice
- Rolling back drivers
- Using different GPUs
- Running Memtest 64 for 3–4 hours (no errors)
- Disabling PBO and similar features
- Checking the motherboard visually for damage (none visible)
- Only one mini ITX PCIe port, limiting testing options
Reliability History and Event Viewer both report: “System Thread Exception Not Handled. What failed: (the driver I enabled be it AMD or Nvidia)”. Apart from this, the system functions normally otherwise.
Could something be related to the motherboard or power supply? The PSU is robust—less than a year old Corsair SFF 750 W Platinum—and I don’t typically play heavily, nor have I had many demanding games recently. My specs are:
Windows 11 Home
AMD 3700X (no overclocking)
Aorus X570i Motherboard
AMD 7900 XTX (or 970)
500 GB NVMe SSD and 250 GB SATA SSD (OS installation for MOBO concerns)
16 GB 3600 RAM
750W Platinum Corsair SFF PSU
Consider using the built-in Windows troubleshooters. They might identify and resolve issues.
Run "dism" and "sfc /scannow".
See guidance on using DISM and SFC for repairs.
Download GPU drivers directly from the manufacturer's site, reinstall and adjust settings.
Avoid third-party tools or installers.
Hi Ralston,
I've also given those a shot. Any driver (AMD/Nvidia) including those from Windows update caused the system to crash right away.
For sfc /scannow, I ran it and found a few errors that the system resolved—just tried it three times with the last two showing no issues—but it didn't work in the end. I also reset the CMOS in a last-ditch effort, but that didn't help either.
Hey everyone,
I hope this isn’t the wrong place, but I’ve never encountered this before in all my PC construction experience. I’m running out of ideas about what might be causing my system to fail.
For background, I recently assembled a new build during the pandemic in late 2020. Everything was brand new except for my GPU—a GTX 970 that I intended to swap out once I found a game I really wanted to play. By early August 2023, while running a particular title, my computer would crash repeatedly. It would restart but remain unstable. I believe the outdated GPU is finally giving up. Whenever I attempt to turn on the GPU driver, the whole setup freezes.
I told myself it was okay since I still wanted to play Starfield. So, I upgraded to a new 7900XTX and performed a fresh Windows 11 installation (complete reset). But the problem continued. When I tried anything other than the default Microsoft driver, the system would crash immediately. I had to go back to a previous restore point or enable safe mode to disable the driver.
Curious about RAM, I ran Memtest 86 and got no errors. I swapped out the motherboard and the PSU, but the issue remained. If I attempt to enable any GPU driver, the system freezes and crashes instantly.
What could be causing this? Any suggestions from others would be greatly appreciated.
Things I’ve attempted:
- Fresh Windows 10/11 installation
- Installing drivers only
- Running SFC /checknow (errors fixed each time)
- Memtest 86 scans (no issues found)
- Motherboard replacement
- PSU swap (same model, new)
- Full system cleanup
- CPU stress test with Cinabench (temper stayed under 60°C)
- Disabling XMP, AMD PBO, etc.
- Placing a small crucifix in the case for luck
System specs:
Windows 11 Home
Ryzen 3700X (no overclocking)
16 GB G-Skill RAM 3600
AMD 7900XTX stock (Nvidia 790)
500GB NVMe drive
250GB SSD
Corsair SFF 750W Platinum PSU
Let me know your thoughts!
Stay in one thread unless you're ready for a brain-eating disaster.