Question Games keep crashing with "driver timeout" -- Radeon R9 380 ?

[Theoria]

UPDATE (MOSTLY SOLVED): Normally, Subnautica, Tunic, Another Crab's Treasure and Solar Ash crash when the GPU reaches 86+ degrees. Finally, I found this: Subnautica and Tunic don't crash if I cap their FPS at 60 with RivaTuner. Another Crab's Treasure does still eventually crash, but it takes longer; it might be possible to prevent it from crashing by disabling all the optional graphics settings, or increasing fan rates. I haven't tried Solar Ash, but I'm very sure it'll be similar. The Witcher 3 crashing seems to be a mostly separate phenomenon. Now, whether these facts point to software failure, hardware failure or simply GPU not being strong enough, I do not know.

Some games (Tunic, Subnautica, Another Crab's Treasure) keep crashing. The latter two crash almost immediately. I played Tunic for maybe an hour, and then it started crashing within a minute or so of playing. It doesn't crash with the "driver timeout" notification every time, but it crashes every time.

Also, I've noticed that the problem seems to be worst with 3D Unity games. But not all Unity games crash (Plate Up! and Fall Guys don't crash). Non-Unity games seem fine, I don't remember Civilization VI, Sims 4, Dark Souls II or Dragon Age: Origins crashing in recent memory -- at least not unplayably, like Tunic. But maybe it has nothing to do with Unity and more to do with the recency of the game, since the graphics card is outdated and the last available driver version is from 2021.

Update: I just tried an Unreal Engine 4 game called Solar Ash, and that one crashed too. The report:

Update 2: Okay, apparently Witcher 3 also crashes now (which I played for a hundred hours without issue years ago). I guess something's deeply wrong. Civ VI doesn't seem to crash so far. The Sims 4 doesn't either.

I've tried (partly based on the suggestions on this thread):
1- Lower settings, turn off V-Sync
2- Run "sfc /scannow" to catch and fix any possible DirectX corruption.
3- Uninstall all MS Visual C++ variations and install latest one
4- Turn off "processor performance boost mode" on Windows
5- Disable Xbox Game Bar

The Radeon Software does give me crash reports in the form of a .zip file, but I'm not sure what part to share. Here's a link to the whole .zip file on Google Drive.

I'd add a system report by CPU-Z but I'm not sure how to do that. I guess I can link to Google Drive?

Anyone want to take a crack at it? I got a month of Game Pass and I'm really upset that I can't play the games I really wanted to. Thank you very much.

 

[Fix_that_Glitch]

Using another cable to connect the gpu is a cheap enough troubleshooting step. Have you checked the manufactures web site for any newer drivers? Are the operating temps within normal limits when gaming? Have you tried something like MSI Afterburner to try reducing voltage, memory settings etc. to try to stabilize the card? If it does help it would probably indicate your card is starting to fail.

 

[Theoria]

Using another cable to connect the gpu is a cheap enough troubleshooting step.

This might be a little difficult, as my monitor only supports VGA and DVI, and the card's display outputs are 2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2. I guess I could use a different screen (TV?) or try the other DVI port.

Have you checked the manufactures web site for any newer drivers?

I knew this would've needed to be the first step in troubleshooting, but I thought my drivers were up to date, since the AMD Adrenaline software says it is.

However, when I installed a second Windows on my PC to see if anything would change, I realized the recommended driver on AMD's website is somewhat more recent. (I don't know why the one I have installed doesn't update.)

Regardless, the same failure happened on that other Windows, with the newer drivers, too. So that's most likely not part of the issue (or unsolved by newer versions).

Are the operating temps within normal limits when gaming?

They seem to exceed 80 degrees rarely, if even 70, but I also rarely play demanding games. I did a FurMark 1080p benchmark with a max temp of 79 degrees and a score of 2709.

Have you tried something like MSI Afterburner to try reducing voltage, memory settings etc. to try to stabilize the card? If it does help it would probably indicate your card is starting to fail.

I'm very new to all this, and I tried a few things after I already solved the issue by capping FPS. Based on research, and some helping comments on Reddit, I undervolted the card slightly and edited its fan curve (which for some reason you can't do on AMD Adrenaline). This seems to have had a positive effect on temperatures, but I'm not sure, I haven't tested much. I'll run a benchmark with FurMark again and update this post with my findings.

FurMark after 50 minivolts of undervolting and fan curve edit (reaching 100% speed at 80 degrees): 73 degrees max temp with 2986 points as a score. That's a big difference! Or is it? Again, I've never done anything like this before, so I don't know what's considered good or normal.

Thank you very much for your help! I take it this all means my card's simply getting old?