Question Black Screen Crash. GPU or PSU?

[Corls2]

PC Part List

• CPU: Ryzen 5 5600
• RAM: Teamgroup Tforce Dark Za 32GB (16x2) 3600MHz
• GPU: AsRock Phantom Gaming D 6800XT
• Motherboard: MSI B550M Pro-VDH WiFi
• PSU: Cooler Master MWE 750W Bronze V2 230V

Problem Description​

I built this PC 4 months ago and replaced my GPU 2 months ago (bought second-hand).

Issue: When booting from a cold boot, the PC crashes every time within 10 minutes. Sometimes the GPU fans go to max speed, or the PC restarts with the VGA light on the motherboard and the GPU fans running normally, but shows no signal. If I force restart immediately after the crash, I encounter a boot loop with the VGA light on, still no signal. I have to turn off the PSU and hold the power button for 10-30 seconds to boot again. Sometimes the gpu drivers also get disabled/corrupted. After several attempts (20-40 minutes), the PC stabilizes and can run for hours without issues, even under load. This issue reoccurs every time I cold boot the next day.

Additional Observations:

• Audio continues to play during a black screen crash
• Not overheating
• No errors detected in memtest Vulkan from github
• SOC Power and Memory Current drop to 0 during crashes, according to HWINFO logs
• Event ID 41 in Event Viewer, no minidumps
• Both 1TB SSDs are at 100% health according to Crystal Disk

Troubleshooting Steps Tried​

• CMOS reset
• Reseated GPU
• Tried different CPU
• Tried different RAM
• Tried different PSU (700W Gold tho)
• Tried various driver versions
• Replaced motherboard

Question​

I'm trying to diagnose whether the issue is with my GPU or PSU, as I plan to replace the faulty component this week.

HWINFO but only has SOC drops
HWINFO but more buggy stats

 

[Lutfij]

PSU: Cooler Master MWE 750W Bronze V2 230V
How old is the PSU in your build?

Tried different PSU (700W Gold tho)
Gold is the advertised 80+ efficiency rating for the unit and it doesn't denote if the unit is a good or bad unit(out of the factory) while 700W is the advertised wattage of the unit. What is the make and model of the PSU? What did it power prior to you build and how old is the unit? You should look into sourcing(borrow, not buy) a reliably built 850W unit and then see if the issue crops up.

Replaced motherboard
What did you replace the board with? Like for like? Did you reinstall the OS after swapping your motherboard?

I built this PC 4 months ago and replaced my GPU 2 months ago (bought second-hand).
If the issue cropped up after dropping in the used GPU, the issue isn't your motherboard, it's your GPU/PSU.

 

[Corls2]

The PSU in my build is only 4 months old. From what I remember, it was the Cooler Master G700 Gold.

I had to replace my motherboard because I bricked my first one while trying to fix this problem by updating the BIOS, which then crashed. My first motherboard was an ASRock B550M Pro4, which I replaced with the MSI B550M Pro-VDH WiFi. I also formatted my PC twice, but it didn't solve the issue.

When I tried booting from hibernate before, it fixed the issue for only 4 days. Booting from hibernate does not work anymore.

I had no problems with the GPU for a whole month after I bought it.

 

[Corls2]

I didn't turn off my pc yesterday to test it, currently running it for 21hrs straight without any problem.

 

[Jason H.]

PC Part List

• CPU: Ryzen 5 5600
• RAM: Teamgroup Tforce Dark Za 32GB (16x2) 3600MHz
• GPU: AsRock Phantom Gaming D 6800XT
• Motherboard: MSI B550M Pro-VDH WiFi
• PSU: Cooler Master MWE 750W Bronze V2 230V

Problem Description​

I built this PC 4 months ago and replaced my GPU 2 months ago (bought second-hand).

Issue: When booting from a cold boot, the PC crashes every time within 10 minutes. Sometimes the GPU fans go to max speed, or the PC restarts with the VGA light on the motherboard and the GPU fans running normally, but shows no signal. If I force restart immediately after the crash, I encounter a boot loop with the VGA light on, still no signal. I have to turn off the PSU and hold the power button for 10-30 seconds to boot again. Sometimes the gpu drivers also get disabled/corrupted. After several attempts (20-40 minutes), the PC stabilizes and can run for hours without issues, even under load. This issue reoccurs every time I cold boot the next day.

Additional Observations:

• Audio continues to play during a black screen crash
• Not overheating
• No errors detected in memtest Vulkan from github
• SOC Power and Memory Current drop to 0 during crashes, according to HWINFO logs
• Event ID 41 in Event Viewer, no minidumps
• Both 1TB SSDs are at 100% health according to Crystal Disk

Troubleshooting Steps Tried​

• CMOS reset
• Reseated GPU
• Tried different CPU
• Tried different RAM
• Tried different PSU (700W Gold tho)
• Tried various driver versions
• Replaced motherboard

Question​

I'm trying to diagnose whether the issue is with my GPU or PSU, as I plan to replace the faulty component this week.

HWINFO but only has SOC drops
HWINFO but more buggy stats

The PSU in my build is only 4 months old. From what I remember, it was the Cooler Master G700 Gold.

I had to replace my motherboard because I bricked my first one while trying to fix this problem by updating the BIOS, which then crashed. My first motherboard was an ASRock B550M Pro4, which I replaced with the MSI B550M Pro-VDH WiFi. I also formatted my PC twice, but it didn't solve the issue.

When I tried booting from hibernate before, it fixed the issue for only 4 days. Booting from hibernate does not work anymore.

I had no problems with the GPU for a whole month after I bought it.

I mean, do you have onboard graphics to test if its the GPU?

 

[Corls2]

I mean, do you have onboard graphics to test if its the GPU?

Nope, I only have the 5600, which does not have an integrated GPU. I'll be picking up an XFX merc 319 6800XT tomorrow from someone I know, and will test to see if it still crashes. If it still crashes when I test it tomorrow with a different GPU, does that mean my PSU is the culprit then?

 

[Corls2]

GPU was faulty