Question Black screen when putting a load on my GPU and using a certain PCI slot ?

[ApplesGoneBloopie]

Specs: https://www.newegg.com/p/3D5-000D-000N7?Item=9SIAHT3EER4794

Within the past two weeks, I've been getting an issue wherein using the top PCI slot of my motherboard (a PCI-e 4.0 x16) has caused me to have a black screen when paired with my 3080. The bottom PCI slot (a PCI-e 3.0 x4 IIRC) does not cause this issue, but it's a severe hit to my GPU's performance.

These black screens force me to shut down my PC, and they are usually paired with Event Viewer errors and/or warnings such as "Winlogon in session 1 (console) reuqested session stop using GPU, returned status STATUS_SUCCESS, with progress stage of successful", "The Desktop Window Manager process has exited. (Process exit code: 0xe0464645, Restart count: 7, Primary display device ID: NVIDIA GeForce RTX 3080)", and Event ID 1000 errors.

I've seen some people share similar issues wherein their GPU fans would start ramping up when the screen goes dark, but I've luckily not had that. However, about half the time when this occurs, my audio dies with it. Sometimes, though, I can still hear through my headset, and I know that my keyboard, microphone, and mouse are all still functional.

It seems that some measure of stress is needed for the black screen to occur. I don't recall getting the issue when I just browse the internet, but playing somewhat intensive games eventually brings the black screen around, and then it occurs with more frequency.

I've made sure to update all of my drivers and apply all Windows Updates. I also checked to make sure that I didn't apply any particular updates right before the issue started occurring.

One weird thing that occurred, though, was that buying a GPU support bracket actually fixed my problem for about a week. I thought that maybe GPU sag was causing this, and while that supposedly wasn't the cause (otherwise the bracket would've worked longer than a week), I don't know why it fixed my problem for about a week. I made sure that GPU sag wasn't the issue by doing some overcorrecting, and sure enough I still got the black screen.

I do not have another GPU to test this with, and I don't really have access to another GPU either.

 

[JohnBonhamsGhost]

buying a GPU support bracket actually fixed my problem for about a week....I made sure that GPU sag wasn't the issue by doing some overcorrecting, and sure enough I still got the black screen.

whether the PCIe slot was damaged during production, transport, or later installation can lead to issues like you've seen.
it's possible that your heavy card just broke the camel's back.

using the support may have corrected the angle of connection for a short time but if it was already damaged then it could have just later fully lost the ability to properly connect.

either confirm malfunction with troubleshooting on your own by swapping in\out other hardware or contact the manufacturer now and file for RMA for this board.

 

[ApplesGoneBloopie]

whether the PCIe slot was damaged during production, transport, or later installation can lead to issues like you've seen.
it's possible that your heavy card just broke the camel's back.

using the support may have corrected the angle of connection for a short time but if it was already damaged then it could have just later fully lost the ability to properly connect.

either confirm malfunction with troubleshooting on your own by swapping in\out other hardware or contact the manufacturer now and file for RMA for this board.

I think you're definitely correct that I should contact the manufacturer and swap in/out other hardware. However, what confuses me is that this doesn't happen when my GPU isn't under load, but it's almost certainly tied to the PCI given that switching to the inferior PCI slot still allowed me to play games albeit at significant framerate losses.

At the moment, my GPU is in the potentially faulty slot, and I've been online for a couple hours without an issue. I've also been playing some League of Legends, and the game isn't intensive enough to make this issue occur.

Also, just before switching to the potentially faulty slot, I applied a BIOS update that the B550M-C support page did not recommend.

I'll edit this post/reply with a new post if the same issue occurs again with the BIOS update. Thank you for replying.

 

[JohnBonhamsGhost]

I applied a BIOS update that the B550M-C support page did not recommend

they have an updated BIOS available but they recommend you do not use it?
that wouldn't make sense unless maybe it was geared towards a different series of CPUs.

what confuses me is that this doesn't happen when my GPU isn't under load

have an OSD running showing the GPU power usage.
if you notice higher usage as the issue(s) occur then it's definitely related to the load.


it's also possible that you don't have an adequate power supply for running this card @ 16x under full load.
and possible that you've got a malfunctioning graphics card that can't handle that stress.
to verify on your own you'd need another graphics card
and another power supply with adequate wattage for that card.

include your complete system specs in the thread, not just a link to chase.

 

[ApplesGoneBloopie]

they have an updated BIOS available but they recommend you do not use it?
that wouldn't make sense unless maybe it was geared towards a different series of CPUs.

have an OSD running showing the GPU power usage.
if you notice higher usage as the issue(s) occur then it's definitely related to the load.


it's also possible that you don't have an adequate power supply for running this card @ 16x under full load.
and possible that you've got a malfunctioning graphics card that can't handle that stress.
to verify on your own you'd need another graphics card
and another power supply with adequate wattage for that card.

include your complete system specs in the thread, not just a link to chase.

I think they didn't recommend the BIOS update because it resets all the settings in the BIOS? Dunno.

I'll definitely look into an OSD if the issue happens again.

Ryzen 5600X, B550M-C motherboard, ZOTAC (oof) 10GB VRAM 3080, TEAMGROUP-UD4-3200 2x16GB, and an 800 watt 80+ gold PSU.

 

[JohnBonhamsGhost]

think they didn't recommend the BIOS update because it resets all the settings in the BIOS? Dunno

every BIOS update you do, no matter the board or the revision, resets BIOS settings to defaults.

an 800 watt 80+ gold PSU

what exact make & model?
and how long has it been in use?

 

[ApplesGoneBloopie]

every BIOS update you do, no matter the board or the revision, resets BIOS settings to defaults.

what exact make & model?
and how long has it been in use?

Sorry for the late reply. The PSU is an Aperia 800W Prestige 80+ Gold Power ATX-PR800W. I've been using it nearly daily, and for long periods of time, since roughly 8/25/2021.

The randomness of my issue is really confusing me. Sometimes, I can be playing games for something like ten hours and be fine, including games that have given me the black screen, and sometimes it seems to do it instantly. I've even tried to replicate the circumstances and have not gotten the black screen.

Since my last post, I've turned off hardware acceleration on everything I can except my iCUE software because it enables my mouse to work properly. I've gone through plenty of Windows Event Viewer issues and applied fixes. I reinstalled nearly every driver on my PC. I did more cleaning including cleaning my PC literally as well as stuff like junk file removal and virus scanning. I did a complete uninstall and reinstall of everything related to my graphics drivers including any Intel drivers I randomly had as well as many of my AMD chipset drivers. Lastly, I've unplugged almost everything in my PC, cleaned it, and then replugged it.

With every new fix I attempt, I can never tell if I'll be good to go. Sometimes I go a day without a black screen, sometimes I go three hours, and on a couple of occasions after a black screen recently occurred, I've only gone 15 minutes. I can think of some consistencies such as the more stressful games, the PCI slot, and some softwares being used, but I still can't put my finger on anything to be certain.

I'll look into an OSD now since I got another black screen.

 

[yoname21]

The randomness of my issue is really confusing me. Sometimes, I can be playing games for something like ten hours and be fine, including games that have given me the black screen, and sometimes it seems to do it instantly. I've even tried to replicate the circumstances and have not gotten the black screen.

Were you able to find out how to fix this? This problem sounds exactly the same as mine, I put a support in and it fixed the issue for 3 months but the black screen issue is happening again. I've even tried memtest, intel processor test tool and everything looks ok so at this point I'm thinking either the graphics card is faulty or the PCI slot is bad.