Restart with a Black Screen
Restart with a Black Screen
Approximately two weeks ago my machine began displaying a complete black screen before restarting. This happened before with what seems like temporary power fluctuations, which resolved after upgrading my power supply unit. Now I’m confused because my 1000W PSU should be sufficient, and this issue occurs even when the system isn’t under load or idle. To make it clearer, the problem isn’t related to GPU driver failures that would reset automatically; instead, the whole PC shuts down completely, the motherboard lights change through red, green, and white stages, and the computer restarts on its own. The fans cease spinning. I’ve tried various fixes listed below along with my system details.
Processor: 9070 XT
GPU: Hellhound 7800X3D NH-D15
RAM: 32GB
CPU Speed: 6000MHz
Graphics Card: Gskill Flare
Display: Asus B650E-wifi EVGA 1000W gold
Steps I attempted:
- Changed power settings from performance to balanced
- Applied the curve optimizer on CPU from -30 to -25 (still crashes quickly)
- Reduced RAM to 5800MHz
- Updated from Windows 10 to Windows 11
- Performed a Mdsched RAM test – no errors found
- Reseated GPU and all PSU cables
- Disconnected the computer from a power strip and plugged it directly into the wall
Event viewer isn’t providing any clues about the cause, and there’s nothing near the crash that suggests it’s triggering the failure. I’m stuck and would appreciate guidance on troubleshooting next steps.
Have you checked the stress on RAM, CPU and GPU? Also, using the curve optimizer at -20 seems reasonable.
So I've had this setup for months now, bought this 9070XT on release date and it hadn't been doing this using all these changes from before. I will attempt to remove the undervolt from curve optimizer and restore to default in BIOS, but yeah it just seems like it likely isn't from that given it happens at idle sometimes. I've ran the AMD Adrenaline Center stress test for 3 minutes, it failed the first time, I then reseated the GPU it passed. With that being said it then crashed again after reseating so... that wasn't it. I do believe it happened around the time I watched this video: With that being said nothing in here seems like it would cause any issues... I even changed the things back in here that I did. Obviously I didn't do the Nvidia settings since this is an AMD card.
Switching to disable PBO and turning off the curve optimizer. I'll let you know if anything goes wrong.
It’s probably a BIOS or AGESA problem instead of PSU or temperatures. First, update the BIOS for your Asus B650E. Then run everything in stock mode without any curve optimizer or RAM tuning. Perform a stress test using OCCT power and memory to check stability. If it works at stock with the new BIOS, the issue was likely memory or curve settings. If it still fails, the RAM kit at 6000MHz might be the culprit—some Flare X5 models have compatibility problems. Consider switching to a 5600MHz EXPO instead.
If it continues to behave at 6000, definitely try 5600 EXPO. You might also run a test with all settings off except EXPO once, to confirm whether the RAM is the issue.
You haven't tried OCCT yet for testing—consider exploring its documentation and setting up a controlled environment to evaluate its performance.
In OCCT you can execute the Power test and check the box that also highlights Memory. This ensures both CPU and RAM are tested together near their maximum usage. Start by running it for 30 minutes initially, then extend to 1–2 hours if successful. Maintain HWInfo open in the background so you can monitor temperatures and voltages during testing. If it crashes here, it indicates a stability problem with RAM or CPU settings rather than just gaming load.