Death of a graphics card
WendyLuvsCatz
Posts: 38,933
before I hold the funeral I am still trying to rule out other reasons for my 980ti crashing and never recovering
it started 18 of July with a big Windows update and I tried clean driver installs, roll backs of updates and all sorts of stuff
it has persistently gotten worse
it was only crashing rendering with GPU based programs
now
it crashes opening the start menu!
I can use my PC in safe mode only in effect
I will try my old 750 tomorrow to rule out any other cause but it is likely my 980ti has carked it by looks 
am mainly posting to see if others who have lost a dearly beloved card can say if this is how it went down
share some sad memories so I can know I am not alone and what told you it was the end
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2025 Daz Productions Inc. All Rights Reserved.
Comments
If everything works fine in safe mode it's probably not the card.
Any horrible screen artifacts? If not, then it might not be your card.
No artifacts and when it works it works beautifully
with this happening intermittently
It loved the Uningine Heaven stress test rendering
crashed doing mundane stuff
happily rendered scenes using all 6GB in Octane, iray etc and when it did black screen iray kept going and rendered the rest of the frames on the CPU according to the DAZ studio logfile
But now I get black screens almost immediately using my PC doing anything
My PC doesn’t crash though
just no signal to my monitors
iray kept rendering frames in CPU mode afterwards when I was using DS one crash too
Are you checking temperatures on the card? It's possible that one or more fans have died and the card is overheating.
Not overheating
I run GPU-Z all the time
it happily sits at 60C most of my renders
can get to 75C gaming
anyway now it’s crashing minutes after startup just opening stuff
I never had a 980ti go down in fact i still have 2 f them since i upgrade to 1080ti's a few months back.
But I have had 2 other cards in the past croak. A 680 & 970SC went the way of the angels the first thing i noticed with both cards down is windows was using the windows on board graphics which is terrible, the icons are huge the screen is very pixilated, etc. I had better graphics in safe mode. when my cards went down it was quick, no warning signs and every time when i started windows. I could only use the windows onboard graphics .
It may not be your graphic cards Wendy. Have you tried a windows restore? windows updates can play havoc with a PC's sometimes. if you have another graphic card you can test it to see if its not something in the integrity windows to see if it was your card or not . usually though when a graphic cards dies , windows falls back on the on board graphics If it turns out to be a PC windows update I hope you kept a recent back up of your pc before the update. If it was me I would hope its a graphic card issue that would be a easier fix. I'll keep my fingures crossed for uo Wendy
Check your RAM, bad RAM can cause all kinds of strange problems, and they typically include random crashes.
Windows has it's own memory diagnostics tool which runs during boot, I'm not sure how to activate it on the different Windows versions but search on "windows boot memory test windows [version]".
Or use Memtest86+, it's more advanced and runs repeated test cycles until you stop it, windows memory test is fast and may not catch all types of errors. Runs from a bootable CD or an USB key, there are installers available that can create both: https://www.memtest.org/
I ran a memory test and tried restoring old updates
been down all these paths to no avail
now trying a third monitor on the integrated graphics on my motherboard
in BIOS enabled it so if it happens again will see if that one still works
Hmm, you sure it's not the PCI port it's docked to hon?
live crash video
and that third screen useless cannot access anything
I only have one available PCI port because I broke my other one ripping out my 750 a few years ago
Well I created a shortcut to shutdown.exe -r -t I labeled restart and plonked on my third monitor
but it doesn’t stay there restarting
well didn’t I managed to make it stick now
but it does shut my PC down so need to first thing drag it around to that monitor I guess
just textured view in DAZ studio crashes it so I need to do something
either set up my intel graphics and render CPU or try my 750
not many options left now
https://www.nvidia.com/en-us/geforce/forums/geforce-graphics-cards/5/212834/for-all-people-with-black-screensno-signal-issues/
might try some of these though
Shutdown
check your system for any air flow that is blocked or not properly venting because of a air path that's not supposed to be there (e.g. expansion slot cover missing)
Make sure your PSU is getting adequate airflow. You have pictures of cats, do you have cats? Get some compressed air and clean your PSU fan.
Remove all devices except mouse and keyboard, disconnect all but one single display.
plug your PC directly into an outlet by itself and not a power strip
now boot and see how it behaves.
If it still blows out you may want to look to see if your CPU is properly seated and neatly pasted, nothing should be dripping over the edge like a rootbeer float.
...had that happened with the old GTX 460 that I was still using to drive just the displays. Replaced it with a 560 Ti I have and everything works fine.
Have you swapped out your 980 and 750 yet? You said you broke your other PCIe slot removing a video card? Did you pull the frame off and leave all the connectors exposed?
I am with you Wendy I think its your graphic card too. after you posted your demo that is exactly what happened to both my failing gpu's, the only reason I believe its your gpu is because your monitor with the on board graphics enabled in bios was still running, that sucks about the card, but at least it should not be hard to fix if you can get one cheap enough, maybe find a refurbish one to get you by with until you can afford something better.
I was going to come back tell you i just had to do a system restore on my neighbors computer after she had a windows 10 update she could not get windows to run her graphics card, but we did a windows/restore and updated her driver for her gpu and it seem to solve her issue . so I with you i think your gpu is in the process of going the way of the angels . that really unfortunate
GTX 1060 6GB (used) cards are going on ebay for below $140.00 Just picked up one for a friend.
That it instantly crashes as soon as any load is put on the GPU is a very bad sign. That pretty much means its either the GPU, the slot or the traces on the motherboard.
Since you have already one bad slot I'd lean toward the motherboard being bad but the only real way to tell is to put the card into s system that is otherwise working and test it, or vice versa, a new good card into that system.
Yes connectors exposed, the plastic bit was ripped off, this was 4 years ago and the motherboard was new
if need to replace it I might as well replace the PC
It could also be the VRM going bad or even some capacitors giving up the ghost.
Also, the on-board temperature probes may be accurate. Or maybe not. Yeah, it's like that.
Good luck, but the 1060 prices do look very nice.
Let us know when you plan to hold the viking funeral for the old card.
I did that once. The connectors got all tangled up and I couldn't even boot the thing after that. :(
Sent you a PM.
its down to windows being rubbish and installing the OEM display driver, while then you need to install the nVidia driver again from their site, and the clean install within the installer is rubbish, so you need to download and use DDU from https://www.guru3d.com/files-details/display-driver-uninstaller-download.html to fully uninstall driver's and start over
I did completely remove the driver so that windows even thought my card was new hardware
and all the older installs about 20GB of them, I then treated it as new hardware and when given options to get a driver had my Nvidia one ready and it worked a few days.
yesterday reinstalled the driver with GeForce Experience too
Windows updates certainly coincide with this crap were big ones yesterday and now it’s completely cactus
I had an old video card a long time ago which ran perfectly fine until I tried to play any game. The card was factory overclocked, so I tried setting it to default clocking and the VC worked perfectly after that. Have you tried under clocking your VC?
Also, in 2016 I bought an MSI GTX1080 sea hawk (the one with water cooling), it had minor stability issues pretty much from the day I bought it. Crashes got more frequent over time until it crashed basically every time I tried to render with Daz Studio. 2.5 years after I bought it, the thing stopped working entirely. Fortunately, it came with a 3 year warranty and I got a full refund which I put toward buying an RTX2080ti instead.
Not the VRM's. They supply power to the CPU and if one was going bad the entire system would crash not just the GPU.
Cap's on the MoBo are also unlikely as if they are bad then no signal would cross them and that wouldn't entirely crash just the GPU.
Videocards have their own VRN, along with caps.
This looks like a TDR crash, which is not actually a crash crash but Windows or the videocard needlessly disrupting the videosignal because it thinks there's something wrong; can be caused by anything from bad software updates to a fan halting it's spin (either because of dust or a technical fault like low power or broken parts; this does not always show up in monitor programs, physically check it during crash!).
So 3 tips before spending money on a new card:
- update or install older updates from when it worked (you can deinstall windows updates too. Some are just bad, like there's this one infamous Windows update that 100% kills my pc everytime I try to install it, but it works fine for 99.9% of the rest of the world, so Microsoft ain't fixing it.)
- check the fans for sudden slowdowns, halts and dedust them (if it's a faulty fan, you can sometimes replace them for like 20 dollars and a screwdriver)
- dig into TDR settings in Windows. It's possible an update set them wrong and you need to delay the timer and everything is fine again (had that once too after an update, thought my videocard was freaking out everytime it needed to do anything, turns out it was just the dang software causing it)
mine is Win7
I tested not using DPI or HDMI to no avail and installing with only one monitor as suggested on the Nvidia forum
it crashes as soon as I do anything that uses the graphics
I will try the 750 later but dreading fiddling inside my rig as that did not go well that time I ripped that PCE lane out
am still trying a few more less invasive suggestions
Has two less pins but two on the plug are move able so hopefully that’s why
I think I know what happened
found my missing METAL air compressor nozzle wedged between the card and the motherboard
I likely shorted something
hopefully not the motherboard