[SOLVED] EVGA 1080 Ti crashing in games. Should I RMA?

[rex4748]

I built my computer in early June and up until a few weeks ago, I've had no issues with it. Suddenly, all my games started crashing and I get this error in the event logs: "Display driver nvlddmkm stopped responding and has successfully recovered."

I looked up possible solutions, and I believe I've tried them all:

- Uninstalled GeForce driver -> Ran DDU -> clean install
- Reformatted and re-installed Windows
- Underclocked GPU
- Changed power settings to "maximum performance", etc. Tried tweaking voltages as well.
- Ran MemTest. RAM is all clear.
- Stress tested CPU. All clear. Tried reverting to stock settings as well.
- Cleaned dust out of computer (there wasn't much) and reseated GPU.
- Tried a different PCIe slot.
- Tried a different PSU, as well as different cables.
- Tried a different monitor and different cables.
- Detached all peripherals and clean booted Windows.
- Tried disabling G-Sync, and tested just about every Nvidia setting
- Tried playing games in Windowed mode

The airflow in my case is fine as well, and my temperatures are at reasonable levels at all times (under 70°C). I've tried this with several games and they all crash within half an hour or so (average is about 5-10 minutes). Like I said, I've had no issues running these exact same games for the past 4 months. This just started out of nowhere; I hadn't updated or changed anything. I'm really confused as to why the card would just start doing this months later if it was defective, rather than right away.

Anyway, I can't think of anything I haven't tried, and at this point it seems like my only option is to RMA the card.
Just wanted to make sure I wasn't missing anything before I go ahead with it. Any thoughts or suggestions are appreciated!

Specs:
ASRock Z370 Taichi
i7-8700K
EVGA 1080 Ti SC Black
32GB RAM (G.Skill)
Samsung 860 EVO SSD
Corsair RM650i
Windows 10 Enterprise (64-bit)

 

[Barty1884]

Have you tried a different/older driver? Or are you removing then reinstalling the same one each time?

While it's possible for a card to 'go bad' after a few months of use, they're typically either DOA, or will last years (under 'normal' circumstances).

No harm in RMA'ing; better safe than sorry.... But I'd definitively rule out drivers first of all.


Also ensure you're set to max performance, both in Windows & in the Nvidia control panel.

 

[rex4748]

I tried an older driver (v388.13). The one I was on previously was working fine for 4 months and I had not updated it when it started crashing on me. Tried the newest one as well. Crashed on all three. Max performance is on in Windows and in the Nvidia control panel.

Yeah, it does seem odd that it was fine and now it's just not. The card didn't die or stop working under normal conditions, all my games just started crashing for no particular reason, and even three different driver versions and a total re-install of Windows didn't fix it. I really don't get it.