F5F Stay Refreshed Hardware Desktop Question My GPU (1080 Ti) fails to function following a blue screen of death during a performance test.

Question My GPU (1080 Ti) fails to function following a blue screen of death during a performance test.

Question My GPU (1080 Ti) fails to function following a blue screen of death during a performance test.

L
LizardDragon
Junior Member
21
11-06-2021, 10:26 AM
#1
Hello everyone! I'm just starting out on this forum and I'm not very comfortable with computers, but I'm reaching out to try to figure out what happened to me.
Over the last couple of months, I've been delaying replacing my graphics card—a KFA2 GeForce GTX 1080 Ti OC DuraPro with 11 GB of memory—but it's still functioning perfectly. To test its limits, I ran a stress test using Userbenchmark. Everything seemed normal until the end, when I encountered a blue screen. After restarting, a white LED appeared on my motherboard (an MSI PRO Z690-A DDR4). When I opened Userbenchmark, I saw a lot of red pixels, screen artifacts, tiny green strips, and my PC would freeze, crash, or refuse to start. Safe mode worked, the graphics card showed up in Task Manager, but it was like trying to draw on the screen with a mouse.
After several tries, nothing seemed to work. A friend brought over an older 1060 card, which fixed everything. We decided to clear all GPU and Nvidia drivers and reinstall everything. Once that was done, we tried again with my 1080 Ti. It booted up correctly and everything seemed fine. I waited a few seconds before attempting to restart the Nvidia app, but it crashed again—every time the white LED lit up. After a few attempts, we gave up and turned to this forum for advice. According to my friend, the test might have damaged the VRAM or something in my card, though he's not certain. I shared all the details, and if you have any questions or suggestions, please let me know! Whether it’s time to buy a new GPU or if my card can still be saved, I appreciate your help.
Thank you for reading this far!

Here are the full specifications of my PC:
- FOX SPIRIT GT-850P V2 850W 80+P POWER SUPPLY
- CORSAIR VENGEANCE DDR4 32GB (4*8) CMK16GX4M2E3200C16
- MSI PRO Z690-A DDR4
- INTEL I7 12700KF
- SAMSUNG SSD 2.5" 850 EVO 250 GB
- KINGSTON SSD 2048G KC3000 M.2 2280 NVME SSD GEN 4
L
LizardDragon
11-06-2021, 10:26 AM #1

Hello everyone! I'm just starting out on this forum and I'm not very comfortable with computers, but I'm reaching out to try to figure out what happened to me.
Over the last couple of months, I've been delaying replacing my graphics card—a KFA2 GeForce GTX 1080 Ti OC DuraPro with 11 GB of memory—but it's still functioning perfectly. To test its limits, I ran a stress test using Userbenchmark. Everything seemed normal until the end, when I encountered a blue screen. After restarting, a white LED appeared on my motherboard (an MSI PRO Z690-A DDR4). When I opened Userbenchmark, I saw a lot of red pixels, screen artifacts, tiny green strips, and my PC would freeze, crash, or refuse to start. Safe mode worked, the graphics card showed up in Task Manager, but it was like trying to draw on the screen with a mouse.
After several tries, nothing seemed to work. A friend brought over an older 1060 card, which fixed everything. We decided to clear all GPU and Nvidia drivers and reinstall everything. Once that was done, we tried again with my 1080 Ti. It booted up correctly and everything seemed fine. I waited a few seconds before attempting to restart the Nvidia app, but it crashed again—every time the white LED lit up. After a few attempts, we gave up and turned to this forum for advice. According to my friend, the test might have damaged the VRAM or something in my card, though he's not certain. I shared all the details, and if you have any questions or suggestions, please let me know! Whether it’s time to buy a new GPU or if my card can still be saved, I appreciate your help.
Thank you for reading this far!

Here are the full specifications of my PC:
- FOX SPIRIT GT-850P V2 850W 80+P POWER SUPPLY
- CORSAIR VENGEANCE DDR4 32GB (4*8) CMK16GX4M2E3200C16
- MSI PRO Z690-A DDR4
- INTEL I7 12700KF
- SAMSUNG SSD 2.5" 850 EVO 250 GB
- KINGSTON SSD 2048G KC3000 M.2 2280 NVME SSD GEN 4

W
WildCandy
Senior Member
675
11-06-2021, 01:08 PM
#2
The card is 8 years old at this point they do die eventually. Not much more you can do outside of stripping it to see component damage for VRAM
Just make sure all the power leads are seated correctly
W
WildCandy
11-06-2021, 01:08 PM #2

The card is 8 years old at this point they do die eventually. Not much more you can do outside of stripping it to see component damage for VRAM
Just make sure all the power leads are seated correctly

E
Energyx
Junior Member
35
11-08-2021, 03:31 AM
#3
Welcome to the forums, newcomer! I turned off my PC and a white LED illuminated on my motherboard (an MSI PRO Z690-A DDR4). You're referring to the VGA debug LED on the motherboard? Your manual doesn't specify the color for each debug LED. Attempt to capture a screenshot from your desktop to check for consistent issues. Use GPU-Z to look for any zero values on the first page. Then, visit Device Manager to see if your GPU shows error code 43. If you encounter any of these signs, your GPU might be failing.
E
Energyx
11-08-2021, 03:31 AM #3

Welcome to the forums, newcomer! I turned off my PC and a white LED illuminated on my motherboard (an MSI PRO Z690-A DDR4). You're referring to the VGA debug LED on the motherboard? Your manual doesn't specify the color for each debug LED. Attempt to capture a screenshot from your desktop to check for consistent issues. Use GPU-Z to look for any zero values on the first page. Then, visit Device Manager to see if your GPU shows error code 43. If you encounter any of these signs, your GPU might be failing.

S
Skeetarr
Member
146
11-14-2021, 10:02 AM
#4
Begin by checking the Reliability History/Monitor and Event Viewer for any error messages, warnings, or informational entries recorded just before or at the moment the GPU ceases functioning.
Swapping the GPU is beneficial since errors should appear and vanish based on which GPU is active.
Next, identify the exact Nvidia application that requires restarting.
Have you attempted to manually download and reinstall it from Nvidia’s official site? Avoid third-party tools or installers.
Once the system is operational, execute "dism" and "sfc /scannow" to let Windows repair any corrupted files.
I’m not a full advocate for benchmarks or user testing and remain cautious about such procedures.
For clarity and to rule out issues, uninstall any unknown applications and verify no other processes are running in the background.
This process might simply point to an outdated, malfunctioning GPU.
S
Skeetarr
11-14-2021, 10:02 AM #4

Begin by checking the Reliability History/Monitor and Event Viewer for any error messages, warnings, or informational entries recorded just before or at the moment the GPU ceases functioning.
Swapping the GPU is beneficial since errors should appear and vanish based on which GPU is active.
Next, identify the exact Nvidia application that requires restarting.
Have you attempted to manually download and reinstall it from Nvidia’s official site? Avoid third-party tools or installers.
Once the system is operational, execute "dism" and "sfc /scannow" to let Windows repair any corrupted files.
I’m not a full advocate for benchmarks or user testing and remain cautious about such procedures.
For clarity and to rule out issues, uninstall any unknown applications and verify no other processes are running in the background.
This process might simply point to an outdated, malfunctioning GPU.

T
tamemarco
Senior Member
482
11-14-2021, 01:01 PM
#5
It seems like the GPU or power supply unit might be involved. Performance benchmarks can help with overclocking and underclocking, but beyond that they offer limited benefits.
T
tamemarco
11-14-2021, 01:01 PM #5

It seems like the GPU or power supply unit might be involved. Performance benchmarks can help with overclocking and underclocking, but beyond that they offer limited benefits.

T
Tilzy
Junior Member
38
11-16-2021, 01:23 AM
#6
Thanks a lot, everyone!
I tried several of your suggestions and managed to boot the PC with the 1080 Ti. However, it crashes after just 15 seconds. During that time I captured a screenshot of Gpu-Z, and it clearly shows 0 in the Memory section.
The artifacts aren’t visible in safe mode. I saved a screenshot from Reliability History/Monitor and Event Viewer. It’s worth mentioning the blue screen happened at 9:19 PM on the 16th.
I purchased the PSU two and a half years ago, so I don’t think it’s the source of the issue.
When checking errors with the “broken” GPU, the term GameInput Service appears frequently, such as in this message: “The GameInput Service service terminated unexpectedly. It has done this 6 times.”
I’ve saved all the screenshots I could, but please note I’m French and some images might still show in French.
I ran DISM and found no errors, though sfc /scannow reported issues that were fixed.
I removed everything related to Benchmark, and I don’t see any background processes left.
I believe the graphics card is essentially dead, which just helps me decide what to replace next.
https://imgur.com/a/gpu
View: https://imgur.com/a/gpu-F6abDrP
T
Tilzy
11-16-2021, 01:23 AM #6

Thanks a lot, everyone!
I tried several of your suggestions and managed to boot the PC with the 1080 Ti. However, it crashes after just 15 seconds. During that time I captured a screenshot of Gpu-Z, and it clearly shows 0 in the Memory section.
The artifacts aren’t visible in safe mode. I saved a screenshot from Reliability History/Monitor and Event Viewer. It’s worth mentioning the blue screen happened at 9:19 PM on the 16th.
I purchased the PSU two and a half years ago, so I don’t think it’s the source of the issue.
When checking errors with the “broken” GPU, the term GameInput Service appears frequently, such as in this message: “The GameInput Service service terminated unexpectedly. It has done this 6 times.”
I’ve saved all the screenshots I could, but please note I’m French and some images might still show in French.
I ran DISM and found no errors, though sfc /scannow reported issues that were fixed.
I removed everything related to Benchmark, and I don’t see any background processes left.
I believe the graphics card is essentially dead, which just helps me decide what to replace next.
https://imgur.com/a/gpu
View: https://imgur.com/a/gpu-F6abDrP

D
drewnecros
Junior Member
28
11-16-2021, 01:45 AM
#7
I'm not sure how to respond, but according to Msinfo32 the BIOS version is American Megatrends International, LLC. 1.20, dated 2/24/2022. My baseboard version is 1.0.
D
drewnecros
11-16-2021, 01:45 AM #7

I'm not sure how to respond, but according to Msinfo32 the BIOS version is American Megatrends International, LLC. 1.20, dated 2/24/2022. My baseboard version is 1.0.

R
R3dundancy
Member
70
11-17-2021, 06:43 PM
#8
I've been reflecting on this lately.
R
R3dundancy
11-17-2021, 06:43 PM #8

I've been reflecting on this lately.