Question Random periods of high GPU usage causing stutter/lag after hearing loud pop from PC

sam123bir

Honorable
Nov 25, 2016
21
0
10,510
So I got this PC about 3 and a half years ago, and it has been working fine until recently. Now I have this issue where the GPU essentially gives up for about 5-20 minutes at random times. It doesn't happen often, and I cant replicate it no matter how much stress I put on the PC, it's very random. Some days it doesn't happen at all no matter what I do (heavy gaming or Youtube), other days maybe once, and some days it happens multiple times. It happens at very random times, thus far when watching Youtube, once while playing Destiny 2, and once while mostly idle using the file explorer. During these periods of lower GPU performance, any little thing will cause the PC to lag. Watching Youtube at 1080p lags and makes the audio pop/crackle, games where I usually get 100+ FPS goes down to 15 FPS, moving a single window around the screen is laggy, etc. I can't find a pattern and because of it's randomness it's very hard to troubleshoot. I appreciate all the help I can get to find out why this is happening.

  • After reading many threads all over the net of similar (but not the exact same) issues, I thought it could be a PSU issue. Maybe it wasn't supplying enough power to the GPU. But my inability to replicate these laggy periods even when putting high stress on the computer seems to contradict that. I discuss how I ran a CPU and GPU stress test at the same time while watching Youtube further down. Because of this, I believe it could be a driver issue, but I'm not sure (I got latest drivers). I need better insight from smarter people that got more experience.
  • Currently, I have not attempted to uninstall the GPU drivers using DDU and doing a reinstall of them, but that's something I'm considering.
  • I'm also considering doing a full fresh reinstall of Windows 10.
  • I have also set up GPU-Z with logging, so the next time a lag period happens I'll have some logs of it.

IMPORTANT DETAILS:
It all began with the computer starting to lag heavily while I was watching Youtube (at 1080p). Then I heard a loud pop/crack from the computer. No smoke or anything, just a very noticeable pop. It continued lagging and I heard a few more pops, so I restarted the PC. I have not heard a pop since then, even if the computer has had many laggy periods since.

At two separate times (once while watching Youtube, once while playing Destiny 2), the computer lag became so much that the two monitors lost signal and went black for about 10 seconds. The signal then came back and the monitors were displaying again.

I ran Geeks3D's FurMark GPU stress test once (10 minutes), average FPS was 25. This caused the whole computer to lag a bit. This run was seemingly during one of these periods of lag since two subsequent runs had average FPS at 55 (15 minute tests) without any lag. I also ran Prime95's CPU stress test and this worked fine with no lag. I even ran both stress tests at the same time while watching Youtube (at 2K), and the computer handled it fine with no lag. FurMark was averaging at 55 frames with GPU at 80 Celsius, and CPU was at 95 Celsius at the end of the test (which took 40 minutes total). I also ran the Windows Memory Diagnostic to check the memory just in case, but it found nothing.

During these periods of lag, the GPU's overall performance is significantly lowered. It can barely handle 1080p Youtube videos, it lags while moving windows, and it sits at around 80-100% GPU usage constantly until the laggy period is over. When watching Youtube, the applications causing this high usage is Google Chrome and Desktop Window Manager (both are the legit applications, not malware, I checked). The GPU temperature does not increase much, it usually stays in the range it was before the lag period began. There is seemingly no noticeable increase in CPU or RAM usage either.

OTHER DETAILS:
I have run virus, malware, and adware scans of the full computer using Windows Defender, Malwarebytes and Adwcleaner respectfully, yet they found nothing.
Using latest GeForce Game Ready Driver as of writing (536.40).
Using BIOS version 4007.
Using 2x 1920x1080 144 Hz monitors.
Using Google Chrome as web browser.

SPECS;
[CPU] AMD Ryzen 7 3800X 3.9 GHz
[GPU] Asus DUAL EVO OC GeForce RTX 2070 SUPER 8 GB
[Motherboard] Asus ROG STRIX B450-F GAMING
[RAM] Kingston HyperX Fury 16 GB DDR4-3200 CL16
[PSU] Corsair TX650M Gold 650 W
[SSD] Kingston A2000 1 TB M.2-2280
[HDD] Seagate Barracuda Compute 2 TB 3.5" 7200 RPM
[OS] Windows 10 Home 64-bit

Thanks for all help!


EDIT:
I have now updated the BIOS to the latest version as of writing (5102) and cleared the CMOS.
I have now used DDU in safe-mode to uninstall the Nvidia drivers and then reinstalled the latest ones as administrator.
I also decided to do a thorough cleaning of the PC which removed a lot of dust on the fans and the various filters.
 
Last edited:
You seem to have two more BIOS versions pending update;
I'd advise on jumping to the latest, clear the CMOS and then see if the issue persists.

I have not attempted to uninstall the GPU drivers using DDU and doing a reinstall of them, but that's something I'm considering.
That is what you should do. manually reinstall in an elevated command, i.e, Right click installer>Run as Administrator.
 
Hello!
To troubleshoot GPU performance issues:
Update GPU drivers and consider reinstalling them using DDU.
Monitor GPU temperatures and check for overheating.
Test with a different GPU to identify the source of the problem.
Consider a fresh Windows installation if other steps don't resolve the issue.
 
You seem to have two more BIOS versions pending update;
I'd advise on jumping to the latest, clear the CMOS and then see if the issue persists.

I have not attempted to uninstall the GPU drivers using DDU and doing a reinstall of them, but that's something I'm considering.
That is what you should do. manually reinstall in an elevated command, i.e, Right click installer>Run as Administrator.
Thanks for the reply. I have now updated the BIOS to the latest version (5102) + cleared the CMOS. I also used DDU in safe-mode to uninstall the Nvidia drivers and then reinstalled the latest ones as administrator.

Hopefully this will cease the periods of lag, but we'll see. It might take some time to figure out because of how random their occurrences are.
 
Last edited:
I hope your issues are fixed now but I would not ignore the pop's inside the computer where you said this all started. If this came in for me to fix, first thing check power supply. At this point system still lives. At least throw a power supply tester or volt meter on it to check.
 
I hope your issues are fixed now but I would not ignore the pop's inside the computer where you said this all started. If this came in for me to fix, first thing check power supply. At this point system still lives. At least throw a power supply tester or volt meter on it to check.
Thanks for your input. It's gonna take a while to figure out if the issue is fixed or not since it doesn't happen every day - but you're right, I should probably check the PSU just in case.

Is there a preference between using a power supply tester or a voltmeter? Or would either of them work just as good? I have access to a digital multimeter and can test my PSU first thing tomorrow. I've never done this before, but I'm assuming I should perform the test similar to how this video shows?
 
Hello!
To troubleshoot GPU performance issues:
Update GPU drivers and consider reinstalling them using DDU.
Monitor GPU temperatures and check for overheating.
Test with a different GPU to identify the source of the problem.
Consider a fresh Windows installation if other steps don't resolve the issue.
I have already reinstalled the latest GPU drivers after using DDU.
As mentioned in my post, there is nothing noteworthy of the GPU temperatures when a period of lag occurs. There is no overheating.
And I sadly don't have access to another GPU to try out.

I'm looking to test the PSU soon because of the popping sounds I heard from the PC, but if it shows to be working fine then I guess my only option is to do a fresh reinstall of Windows if the lag comes back.
 
I have already reinstalled the latest GPU drivers after using DDU.
As mentioned in my post, there is nothing noteworthy of the GPU temperatures when a period of lag occurs. There is no overheating.
And I sadly don't have access to another GPU to try out.

I'm looking to test the PSU soon because of the popping sounds I heard from the PC, but if it shows to be working fine then I guess my only option is to do a fresh reinstall of Windows if the lag comes back.
Then I think you need to

Test PSU for potential issues.
Run hardware diagnostics.
Consider a fresh reinstall of Windows.
Check for resource-intensive background processes.
Update BIOS and firmware.
Seek professional assistance if the problem persists.
If it works then please tell me.