Hey!

Thanks to Lugum again for the suggestion of creating a full thread in here. I hope someone can help me figure this out as it's been making games unplayable.

So, here's the story:

About 2 weeks ago I started getting blackscreens while gaming, usually lasting 1-2 seconds, and on both screens at the same time. This causes some games to freeze and crash, others eventually run on. Open programs on the secondary screen (mainly Discord and Steam) become black. Still interactable, but I can't see what's happening. My browser stops responding visually but you can do stuff, just not see what it is. Once it starts working again without a restart, the link bar goes black whenever you type anything in there. Only fix is to restart the programs each. I updated the BIOS, the chipset drivers for MB/CPU and Windows itself. I used DDU in safe mode and reinstalled the NVidia drivers. So now I'm pretty much out of ideas what else it coulb be. I did a scan using HWMonitor and saved the logs of the time frame where the issue appeared.

I'll list my specs and add my systeminfo file aswell as the logs that HWMonitor spat out. Links are mediafire, should be alright. If not, let me know, please.

CPU: AMD Ryzen 9 3900X

GPU: Nvidia RTX 3080 (Gigabyte, non-overclocked)

SSDs: Samsung 970 Evo Plus NVMe PCIe M.2 500GB & Samsung 850 Evo 250GB

HDDs: WD Blue 3TB (2015) & Toshiba DT01ACA100 1TB

RAM: Corsair Vengeance LPX DDR4 3200 C16 2x16GB in DIMM B1 and B2

MB: Asus ROG STRIX B550-F GAMING (WI-FI)

PSU: Enermax Revolution D.F. 850w Gold

OS: Win10 Pro 21H1

system information
HWMonitor logs

Thank you very much in advance to anyone who can help me find a solution or contribute their knowledge! Any help is greatly appreciated.

EDIT: I already removed the card and replugged it. Tried different DP and HDMI cables and different ports on the GPU. Fans and temps are okay under load.

EDIT2: I saved the event viewer logs, maybe these can give some clearance? I don't understand a lot about them.
eventviewer

2 years ago*

Comment has been collapsed.

I have you try replough video card, idk if recently clean it maby try to replough it

2 years ago
Permalink

Comment has been collapsed.

Yes, I've done that. The PC has last been cleaned in April but it's hardly collected any dust as it's been built in Nov 2020 and not ran much. Replugged all the cables, including the PSU connection.

2 years ago
Permalink

Comment has been collapsed.

would try to total replough the video card, got last time as well with my video card

2 years ago
Permalink

Comment has been collapsed.

That's what I did. Removed the whole thing out of the PCIe and replugged all cables. The problem persisted

2 years ago
Permalink

Comment has been collapsed.

How's the fans running during heavy load? Any overheating issues?

Remove the card completely and have a look at the pci-e connector for potential damage. Or if it's just not fitted correctly.
Place it back in the comp.. Leave case open and monitor fans and temp during heavy load for 30 min or so.

Keep HWmonitor on one screen.. And your most demanding game on the other.

2 years ago*
Permalink

Comment has been collapsed.

Fans and temp are alright. around 1620RPM and 60°C

2 years ago
Permalink

Comment has been collapsed.

Tip maybe, any seemingly useful answers such as these, edit and add it to your op, else you might get people asking the same questions, and the things you did try like.

[edit]
I tried a.
i tried b.
i tried c. etc

2 years ago
Permalink

Comment has been collapsed.

Could've thought about that, thanks for the advice :D

2 years ago
Permalink

Comment has been collapsed.

:) Don't forget to add your PSU. ;)

2 years ago*
Permalink

Comment has been collapsed.

Gotcha! :D

2 years ago
Permalink

Comment has been collapsed.

What does the Event Viewer say?

2 years ago
Permalink

Comment has been collapsed.

To be completely honest, I have no idea how to use that. I only have the HWMonitor logs so far ans those have some weird bits in there as far as I can tell.

2 years ago
Permalink

Comment has been collapsed.

https://www.howtogeek.com/123646/htg-explains-what-the-windows-event-viewer-is-and-how-you-can-use-it/

(if your windows 10 is german then it might be called different).

2 years ago
Permalink

Comment has been collapsed.

Open Nvidia Control Panel, go to Manage 3D settings, under Global Settings enable "Prefer maximum performance" in Power management mode.

2 years ago
Permalink

Comment has been collapsed.

It's like this already. Is that default?

2 years ago
Permalink

Comment has been collapsed.

i know your cpu doesn't have integrated graphics, but do you have another gpu you could try using, to try to isolate whether it's your current gpu that's causing the problems?

2 years ago
Permalink

Comment has been collapsed.

This is still on my to-do list but might not be possible too soon, as the only other GPU I might have available is my brother's and he's playing games literally all the time and there's hardly ever a minute when he's not home while I'm home because I work more. Will try that as soon as I get the chance though.

2 years ago
Permalink

Comment has been collapsed.

Deleted

This comment was deleted 1 year ago.

2 years ago
Permalink

Comment has been collapsed.

Thanks a lot for the detailed response! HDMI handshake seems alright and it also happens as long as I only use one monitor. My current suspicion is that the GPU is broken in some way or another and it'd be a pain in the arse to get a replacement, even though I have warranty on it. Reinstalling Windows is more of a final-destination solution as I only ever have like 3 hours or so per day to spend on doing anything with the PC, and for some reason in the past this always made things weird, but back then I was even less knowledgeable about computers than I am now. Do you know, by any chance, if the MB chipset drivers might be able to cause this kind of issue? It's been a pain to even update the BIOS using the ASUS AI Suite (the update tool only came with an outdated installation that I had to apply over the recent one, fun times :D) and the MB chipset is separate from the CPU chipset as far as I know

2 years ago
Permalink

Comment has been collapsed.

Deleted

This comment was deleted 1 year ago.

2 years ago
Permalink

Comment has been collapsed.

Audity2000 has some good points.
And you do too about chipsets.
Blacking out makes me think your computer, cable, and monitor are not on the same page. They stop and re-sync.

I just came from buying new "tested" 8k HDMI cables for our new 8k TV as the ones I had were old and slow.
Made a big difference for our TV.

Gaming wise for my computer and monitor, I went with Display Port over HDMI as that was about 5 years ago and Display Port had higher bandwidth = higher refresh.

https://www.pcgamer.com/displayport-vs-hdmi-which-display-cable-should-i-use-for-my-monitor/

If you can take your computer to a friend, anywhere, and try their cable and monitor. I think that would be worth the effort.

2 years ago*
Permalink

Comment has been collapsed.

Yeah, that's something I hadn't considered. He should be using quality VESA-certified cables, and DP whenever possible.

2 years ago
Permalink

Comment has been collapsed.

Thank you for the suggestions, hadn't considered better cables yet. If this was the issue, theoretically, should it keep happening when switching out the GPU with another one?

2 years ago*
Permalink

Comment has been collapsed.

Have you checked the Windows Event Log (or whatever its called, I am not on Windows ATM so can't check) if anything weird/unusual shows up there with the time stamp of the black screens?

2 years ago
Permalink

Comment has been collapsed.

Yet to do that, the Event Viewer was completely new to me until yesterday ^^' I'm in no way an expert, but I'll post anything it spits out as soon as it happens again

2 years ago
Permalink

Comment has been collapsed.

Aand me again, added the event viewer files to the original post at the bottom, maybe this helps identify the issue.

2 years ago
Permalink

Comment has been collapsed.

Deleted

This comment was deleted 1 year ago.

2 years ago
Permalink

Comment has been collapsed.

Thanks :D
sfc didn't spit anything out that seemed noteworthy, might check again.
According the HWMonitor logs the temps and voltages are alright, though the idle voltage of 0.725 seemed a little odd to me, but I'm not too experienced with this kind of thing.

Will have to look into MSI Afterburner, thank you for the suggestion!

2 years ago
Permalink

Comment has been collapsed.

Deleted

This comment was deleted 1 year ago.

2 years ago
Permalink

Comment has been collapsed.

I honestly have no idea what to look out for there. I never overclocked anything and I don't know which of the values afterburner shows are alright or might be worrying. While idling Mem is permanently at about 9230MHz and voltage at zero

2 years ago
Permalink

Comment has been collapsed.

Deleted

This comment was deleted 1 year ago.

2 years ago
Permalink

Comment has been collapsed.

dimm b1 and b2?
so you don't want to run dual channel? because then b2 and a2 would be correct. the power of manuals!
well at least you enabled xmp...right?

anyway at least your temps look good.
i would suspect the power supply but then again it should be powerful enough and looks pretty new. did you connect all the power connectors?

the 3080s had some crashing issues at launch when they boosted the clocks too high on some cheaper cards. but that should have been fixed by the nvidia drivers a loooong time ago.

2 years ago
Permalink

Comment has been collapsed.

Power supply is a player.
My only video card and power supply experiences (more than 1) is that I would get visual corruption, freeze and CTD.
Never recovery, yet I would not put it past the hardware developers to implement recovery to push out the return date of warranties.

Just like hard drives silently recover from bad sectors and move potentially corrupted data do available good sectors.
Kicks the warranty can down the road.
Any time SFC fixes something, its either memory issues or hard drive issues.

2 years ago
Permalink

Comment has been collapsed.

So, about the DIMMs.. My mate who helped be build the rig up was dead set on B1 and B2 being alright and there would not be any major advantage in doing it another way. The CPU cooling got in the way so we settled with this. I hope I got it correct but B1 and B2 are the 2 slots that are furthest away from the CPU, right? Those are the ones the RAM is in rn.
Please don't behead me but I have never heard of xmp until now.

PSU is all hooked up correctly from what I can tell, and I think if it hadn't been, issues would have occurred way sooner. The rig hasn't been changed or modified since Nov '20 when we built it except for software updates and stuff.

Also, unrelated, aber ich mag deinen Namen :D

2 years ago
Permalink

Comment has been collapsed.

fanpost ist immer gerne gesehen. ;)

ich will ja nix sagen aber dein kumpel hat dann doch eher nicht genug ahnung um rechner zu bauen.
ohne dual channel läßt du einfach so 10-15% performance auf dem tisch liegen und alles was man machen muss um es richtig zu machen ist 30 sekunden ins handbuch gucken.
https://dlcdnets.asus.com/pub/ASUS/mb/SocketAM4/ROG_STRIX_B550-F_GAMING/G16610_ROG_STRIX_B550-F_GAMING_WI-FI_UM_WEB.pdf
seite 17. dimm b2 und a2 sind die richtigen und das sind der ganz rechte und der zweite von links.
das sollte man schon machen wenn man so viel geld für so einen rechner ausgibt. ganz besonders bei den preisen in den letzten 12 monaten.

köpfe abgehackt werden hier nicht, das machen nur die neuzugänge. ;P
aber: ohne xmp (intel) / docp (amd) läuft dein ram nur mit standard geschwindigkeit. 2666 oder so. das musst du im bios von hand umstellen damit deine 3200 sticks auch wirklich 3200mhz fahren.

alleine mit den beiden sachen wird deine kiste ein gutes stück schneller laufen. aber abstürze sollte sowas nicht auslösen.
hast du auch jetzt das neueste bios drauf? von den beta versionen würde ich persönlich die finger lassen aber ein mal im jahr sollte man schon nach einem neuen bios gucken. ganz besonders wenn man brandneue hardware einbaut. da kann es manchmal probleme mit der kompatibilität geben. leider muss man nach jedem bios update alle geänderten einstellungen wieder von hand ändern. ich hab dafür immer einen kleinen spickzettel in der schublade liegen, damit man nix vergisst. ;)

hast du auch den neuesten nvidia treiber drauf? die rtx 3080 hatten am start ziemliche probleme weil sie zu hoch geboostet haben und dann einige der billigeren karten abgestürzt sind. gigabyte war glaube ich auch darunter. das war letztes jahr (mal wieder) so ein kleiner skandal bei nvidia...

2 years ago
Permalink

Comment has been collapsed.

Okay, also der RAM sitzt nun in den richtigen Slots, danke für die Erläuterung :) War jetzt mein erster selbst gebauter Rechner und man lernt ja immer mit ^^

Ich muss mir dann mal anschauen, wie das mit docp funktioniert, will ja auch keine Leistung ungenutzt lassen. Hab bisher nie wirklich im BIOS rumgemengt.

Im Laufe der Versuche, das alles zu fixen habe ich das BIOS upgedated (letzte stable, keine Beta), the Chipsettreiber und die GPU-Treiber einmal komplett runter mit DDU und dann neu drauf. Auch mit unterschiedlichen Versionen probiert, da mir gesagt wurde, eine bestimmte ältere Version sei stabiler.

2 years ago
Permalink

Comment has been collapsed.

As few people mentioned before I would suggest checking event viewer. This might be a case of driver failure. In such case windows tries to reset the state, but some programs will fail even if drivers become operational once again. In that case look for "Display driver stopped working" or something similar in the logs. I once experienced just that with the same symptoms - monitors going blank for a moment after which some programs stopped displaying things correctly

Did you also verify correct connectivity from the PSU to the GPU? I know that the report shows stable +12V but your PSU seems to have multiple rails varying in power limits, so be sure the GPU is connected to the right ones

2 years ago
Permalink

Comment has been collapsed.

Will report back as soon as it happens again and see what the Event Viewer has to say about it.

I used 2 8pins that came with the PSU, they were identical and nowhere else to be used when we were almost done hooking everything up so I assumed they're for the GPU. Hooked them both up with the first end, the 2 extension ends (or however you'd call those) are loose on each cable.

2 years ago
Permalink

Comment has been collapsed.

Event viewer files have been added to the original post, very bottom. Maybe these give some clearance?

2 years ago
Permalink

Comment has been collapsed.

Do you use XBox controller on PC? Some users report that same issues can occur if controller is connected i.e. here here

was fighting with the same error for almost a month. Did everything including OS reinstall, tested many nvidia drivers. Turned on v-sync, tested options regarding dwm.exe failures. It didn't help.
It turned out, that the problem was caused by xbox pad firmware.
To update firmware on your pad you have to install xbox accessories from microsoft store and update firmware of the pad.

2 years ago
Permalink

Comment has been collapsed.

Hold on a damn minute..! I recently bought a new Gamepad, XBone, wired.. I swear, if this is it, I'm gonna first lose my shit and second I'll do whatever you want because you might have saved me from A LOT of trouble!

2 years ago
Permalink

Comment has been collapsed.

having the discord overlay can do this somtimes if you have discord make sure the overlay is turned off

2 years ago
Permalink

Comment has been collapsed.

Thanks for the suggestion. I never used the Discord overlay and I doubt it'd be the reason here though, since the program itself turns black and the same happens to the Steam client and parts of my browser.

2 years ago
Permalink

Comment has been collapsed.

k mark that off the list then when i had the discord overlay on it was turning my games/programs black gl on finding what it is

2 years ago
Permalink

Comment has been collapsed.

Thank you. Still appreciate the input nonetheless :)

2 years ago
Permalink

Comment has been collapsed.

Have you checked if there is an updated Bios for your GPU?
The manufacturer may have changed something since launch.

2 years ago
Permalink

Comment has been collapsed.

I never fiddled with GPU BOIS, do you have any advice on how to check and potentially upgrade this? I doubt they come with the NVidia drivers. Thank you!

2 years ago
Permalink

Comment has been collapsed.

https://www.techpowerup.com/vgabios/?did=10DE-2206-1458-403F

But I'd recommend you to test the other things first. Update BIOS as the last resort.

2 years ago
Permalink

Comment has been collapsed.

Thanks for the heads-up, and gotcha, will consider the other things beforehand!

2 years ago
Permalink

Comment has been collapsed.

While you state that your gpu isn't overclocked most gigabyte cards come with a factory overclock. From the data your provided yours boosts to ~2 Ghz which is overclock territory. Might be helpful to know which exact model yours is (Xtreme, master, eagle, gaming oc). What you could try is download MSI Afterburner and underclock your card (especially the vram, because the symptoms sound like memory failure to me) and see if the problems go away. In that case you probably got a card that can't handle the factory oc and you have to see how to proceed from there.

2 years ago
Permalink

Comment has been collapsed.

My GPU is the Gigabyte RTX 3080 Gaming OC, and as I just realized that sounds an awful lot like it could be overclocked, eh? :D

I'll report back as soon as MSI Afterburner gives me some results. Thanks for the advice!

2 years ago
Permalink

Comment has been collapsed.

Deleted

This comment was deleted 2 years ago.

2 years ago
Permalink

Comment has been collapsed.

Thank you for the suggestion! So far I tried unhooking the older Toshiba HDD and one screen and the issue persisted. Will switch out the GPU in Wednesday hopefully, see if that gets me somewhere. Either the problem will persist or appear on my brother's rig, as I'll be switching to his 3070.

2 years ago
Permalink

Comment has been collapsed.

sounds like your gpu or your psu is the issue imo. do a benchmark on the gpu and if you get the same issues i would say it is the gpu. also try making a cpu benchmark.

2 years ago
Permalink

Comment has been collapsed.

Hey, thanks for the suggestion. How do I best go about benchmarking my GPU or doing a stress testt to see if it triggers the issue? Is there a program you'd recommend?

2 years ago
Permalink

Comment has been collapsed.

cinebench and 3dmark is good.

2 years ago
Permalink

Comment has been collapsed.

Thank you, will see how this turns out

2 years ago
Permalink

Comment has been collapsed.

Borrow a PSU from someone and try it. Sounds like that's the issue.

2 years ago
Permalink

Comment has been collapsed.

Thanks for the suggestion. I'll switch out the GPU for another one on Wednesday and see what that does, if the problem persists I'll try the PSU aswell

2 years ago
Permalink

Comment has been collapsed.

Np. Also you may get better help from tech geeks at tomshardware and linustechtips

2 years ago
Permalink

Comment has been collapsed.

It could be a case of residual power. While pc is running push and hold power and reset buttons until pc is switched off and still hold for 5 sec, then power on pc. Residual power is cleared that way.

2 years ago
Permalink

Comment has been collapsed.

Thank you for the suggestion! I already did a hard shutdown yesterday to do something else, would that be sufficient?

2 years ago
Permalink

Comment has been collapsed.

What is the refresh rate of your monitor?

Open NVIDIA Control Panel, In the Display section, click on Change Resolution.

Make sure the resolution is set properly then check the refresh rate. Click Save at the bottom of the screen. Exit & see if it helps.

2 years ago
Permalink

Comment has been collapsed.

Thanks for the suggestion! One of them runs on 60, the other on 120. I already tried various things with resolution, vsync, framerate etc. Nothing made a difference, and in all honesty, the card shouldn't struggle with this either in my opinion.

2 years ago
Permalink

Comment has been collapsed.

Sounds a bit like what was happening just after the release of third party 3080 cards. Some of the manufacturers (including Gigabyte) seemed to be using cheaper capacitors and the cards were struggling to remain stable under load.
This article from last October might explain it, or you can search YouTube, there were a lot of videos about it at the time.

2 years ago
Permalink

Comment has been collapsed.

Thank you! I pretty much ruled that out so far because voltage, temp, fan speed and mem usage remain stable no matter the load I put on it so far.

2 years ago
Permalink

Comment has been collapsed.

You been keeping busy. ;) See how many helpful people there are here. :)

That's a really small event view log, but nothing to worry about in that log.

2 years ago
Permalink

Comment has been collapsed.

Are you sure? I think the Desktop Window Manager crashing is exactly what causes the black screens, but who knows what the cause of the crashing is.

edit: also you can just select some entries and save them seperately, which I assume is what this is. Not the complete event log.

2 years ago
Permalink

Comment has been collapsed.

Maybe not, you are right, i jumped the gun too soon on that.

https://docs.microsoft.com/en-us/answers/questions/237700/dwm-crashing-randomly-causing-black-screen.html

https://docs.microsoft.com/en-us/answers/questions/262545/display-driversdwm-crashing-and-freezing-screens.html
"I think I found the culprit. I think it was a dodgy power plan I got from AMD. I changed it to the Extreme performance version and I've yet to have a crash. Will continue to monitor though and will update accordingly."

Since he has AMD (i got Intel, so unfamiliar with it) maybe worth a look there.

Also https://forums.tomshardware.com/threads/dwm-exe-crashes-playing-rocket-league.3691160/ which mentions an xbox controller which someone else already mentioned too here.

Offtopic lol at his computername enragedhellbeast.

2 years ago
Permalink

Comment has been collapsed.

Yes, I only selected those few events because they kepot repeating and everything around them was unrelated and likely unneccessary for this.

I am actually amazed at how helpful and nice people are in here, especially compared to places like the NVIDIA Discord helpdesk.

Funnily enough, I tried that thing with the controller and updated the firmware and that was the first time in 2 weeks that I could have a game session of 2 hours without any issues whatsoever. The issue literally appeared around the time I got a new pad so I was hyped when InSpec brought that up at first. Will see if this actually fixed it, would be so funny yet kinda tragic, and then again also peak Microsoft :D

Well, I found it fitting, as the previous one was called "underpowered behemoth" as some parts outperformed others by miles, but it was what I had :D

2 years ago
Permalink

Comment has been collapsed.

Okay, quick update:

Thank you SO MUCH to everyone who contributed and gave their opinion/advice! I appreciate every one of you <3

The issue seems to be resolved now and it was actually the gamepad's outdated firmware causing the blackscreens and the dwm.exe crashes.

It seems odd to me but then again I don't know much about all of this. I'm just glad this is fixed and I don't need to reinstall Windows or get a replacement GPU, because even though I got warranty on it I'd probably have to wait for months :v

Thank you all again and have a wonderful week!

2 years ago
Permalink

Comment has been collapsed.

Glad to hear your problem is fixed. Also agree that it is pretty wild for a controller to break a window manager. Whoever messed that up needs to step up their programming.

2 years ago
Permalink

Comment has been collapsed.

Yup, it's so odd yet believable because Microsoft, I guess. But I went through so much stuff to fix this and in the end it was faulty periphery firmware. :D

2 years ago
Permalink

Comment has been collapsed.

Wow, that is insane. I would've never guessed my controller unless maybe Event Viewer revealed a driver issue to me. Thanks for the recap.

2 years ago
Permalink

Comment has been collapsed.

Same, it's just so weird that such a minor thing can cause this much trouble

2 years ago
Permalink

Comment has been collapsed.

Good thing it worked, i expected it to be the controller with what i found and it matched with what Inspec experienced and said, akward problem though.

Makes me check event viewer more too in the future, have a wonderful week too.

2 years ago
Permalink

Comment has been collapsed.

Yep, the man definitely deserves a truckload of cookies :D But you and everyone else too, it's been a nice experience on the internet for once which isn't all that common these days ^^

Thank you :)

2 years ago
Permalink

Comment has been collapsed.

Sign in through Steam to add a comment.