I have a 4090 and for two weeks now I suddenly have this weird issue that the screen turns black and the fans start to spin up. This happens when playing games and in desktop, but while playing I can verify that the game keeps running since it reacts to input which I can tell from the sound being played.
The system was running flawlessly and this started to happen the day after I installed the Game Ready Driver 566.03 as well as the WIndows update KB5046633.
After having a look at my event viewer I discovered the Logitech lamp array bug so I turned of Windows 11’s dynamic lighting option as well as the foreground thing and also disabled lamp array’s service. Likewise, I began hunting for other RGB software and disabled that as well, but the crash persisted. (See here as well)
I then continued looking at the event viewer and searching for errors, which also led to me re-discovering the infamous nvlddmkm problem but at this point I had no RGB software running anymore (*).
I also ran sfc /scannow and DISM.exe /Online /Cleanup-image /Restorehealth but no issues were found.
The next thing I tried was disabling Afterburnger and RTSS and removing it from my autostart. Doing so ensures that it was interfering with the Nvidia App as well as not applying the undervolt I had used. (That being said, this problem does not appear to be an undervolting issue since those are different in my experience.)
Then, I used DDU to remove the driver and install the newly released Game Ready Driver 566.14. Interestingly enough, the install resulted in a black screen. So I waited for a few minutes and rebooted my PC and the driver was installed. I have also noticed that it is not using the Nvidia App but Geforce Experience which I kept since I did not have any problems with it.
Sadly, the GPU is still black-screening and I am starting to wonder what is going on. The event viewer is basically empty now aside from the event of me pressing the restart button. Thus, any help would be greatly appreciated.
*) In fact, Logitech’s GHUB is still running but w/o the RGB service. Nonetheless, I have removed it now from the autostart and I will see if it pops up again.
No, the video card almost never exceeds 65°C (sometimes it might go a bit over 70, can’t remember seeing that though)…and as I have mentioned this also happens in desktop mode and currently I am playing Life is Strange: Before the Storm…on a 4090 XD
I’d try rolling back to the previous “known good” driver rather than those of the 566 series; if it doesn’t return to stability it’d be time to chase other witches.
This might be a good idea. The reddit links above state:
So, it might be a good idea to try 561.09 and 552.22. That being said, I think the one I had before updating was 560.94, but I am not sure. I googled and this is the Game Ready driver for Star Wars: Outlaws and I think this was the one I was using.
Anyway, I will downgrade my driver and see if this resolves the issue. (Currently not at home, I will update this thread on Tuesday)
So today I got around playing a video game again - and using my PC - but before I did that I googled my issue yet again and found the following reddit threads:
Some of them point to (yet another) issue of the 12VHPWR adapter and interestingly enough I am indeed using one from CableMod (not the angled one that has been recalled though).
So, I checked that the cables were properly seated on both the GPU side and the PSU side. In fact, I wanted to replug them but I noticed that they were stuck really hard so I just pushed down an all five of them (1 GPU, 4 PSU).
Then, after I started the PC, I started GPU-Z and enabled logging. My assumption here is if the sense pins are freaking out, I would probably see a voltage drop or something like that for the 12VHPWR connector. (Not sure though…)
Anyway, today I was able to use my PC for around 1.5-2h without any issues. I would not say that I have resolved the problem yet either, and further testing is required. If I experience another crash I will try using the vendor’s 12VHPWR adapter and see if that solves it.
I am still wondering, why the driver/Windows update were causing this issue, but it could have been an unlikely coincidence.
Let’s see how things turn out…I will probably be using my PC over the coming days and no matter the outcome I will keep this thread updated and let you know. (Nothing more annoying than finding a thread after a Google search with the very issue one is having only to be closed with “lol, I was an idiot…solved it” without any further statement )
)