Hi All:
 
First-time post here. I hope not to ask any redundant questions. I've searched all sorts of forums and can't seem to find quite the answer. My ultimate question is what to do about display driver crashes (black screens) whether I have a bad card, bad compatibility with other components, or just bad (or at least unrealistic) expectations. It's kind of a long story, so I'll try as close to the beginning as I can. My specs are posted above for quick reference. 
 

Specs

 

 
CPU: Intel 5960X - Have tried overclocks, but have since reverted to base settings
GPU: EVGA - GeForce RTX 2080 Ti XC Ultra Gaming 11GB (once overclocked, but now the default after Windows reformat -- I hope? I didn't flash GPU bios)
RAM: 32 GB (16x2) Corsair Vengeance Pro RGB 3200, but currently clocked by BIOS at 2133 by the MOBO)
Motherboard: ASUS X99-A II motherboard
Corsair RM750x - Brand new -- bought it because of power issues during Flight Sim
NZXT h510i elite case
 
 

Monitor Timeline

 

• Summer 2019: Bought 2080 Ti XC Ultra Gaming a little over a year ago at Best Buy
• Summer 2019-Last Thursday 2020: Played games more or less fine on both an LG 4K monitor and more recently up until last Thursday a 32" Dell 1440p monitor with up to 165hz refresh rate (though I mostly played at 60hz until recently)
• At some point last week 2020: I upped the refresh rate on the Dell Monitor to 165hz; I also enabled Gsync for the display, which worked in Control Panel even though it wasn't listed as compatible; the card is also running a 1080p second display
• Last Thursday 2020: I update my Nvidia Drivers to the Latest NVIDIA Driver
• Last Thursday 2020: Playing Crusader Kings III for roughly 30 minutes -- Display driver crashes; only recourse is to restart computer. Honestly don't think much of it and go to bed; thought maybe CKIII had a problem and I'd figure it out later
• Last Friday I received the 49" Samsung Odyssey (5120x1440, with refresh rate up to 240hz). I figure 240hz is overkill, so set it up by default to run 120hz 5120x1440p. Fire up CKIII in 5120x1440 at 120hz. Things going great for half an hour, then black screen display crash. Have to restart. Now I'm starting to worry. Similar situation in other games now. Destiny 2. Valorant. All at 120hz, gsync, 5120 1440p
• Tonight: I run the new monitor in 2056x1440 at 120hz (emulating, slightly, the monitor that I was using previously -- note I'm not running the second 1080 display);same crash
• Thinking it's GOT to be the driver, because the latest Nvidia drivers were installed and then I experienced the problem; but then I start to wonder if I had messed with the hz of the monitor before that and I'm not sure -- confounding factor

 

SO here's everything I've tried since then with no effect on the crash

(Edited Sept 26, 2020)

• Reseating GPU
• Running two cables to the GPU
• Verifying PSU connections
• Uninstalling the latest Nvidia drivers with DDU and installing the previous drivers 
• Complete reformat of Windows
• Verifying integrity of game files
• Dusting the PCI-e slot
• Moved GPU another PCI-e slot (they are all the same in my MOBO)
• Swapped to old ram where I had no stability issue
• Swapping to default color profile (not NVIDIA)
• Got temps from two sources. NZXT CAM on the Kraken z63 showed 77 when the GPU crashed; Open Hardware Monitor showed 78 when the GPU crashed
• The crash at 76-78 degrees has proven a reliable reproduction in HWMonitor and Open Hardware Monitor, using 3DMark stress tests and CKIII

 

September 24, 2020 (Tonight)

 

Running CK III at 5120x1440 @ 60hz -- no crash yet (about an hour or so). Which has me thinking . . . Maybe I'm just dumb? Maybe my computer, despite what I thought were beefier components, can't handle above 60hz display and that pixel density and high or medium-high graphics settings. Again bearing in mind that I dropped the resolution to 2056x1440p and ran t 120hz. But I read so much conflicting information on various forums that part of me wonders if I should RMA the card. I also still can't get over the fact that my problems started around the time of the driver update. 
 
And yes I know I don't need to run CKIII at 120 hz, but it might be nice to run Destiny or Vaolorant at that rate. Those also crashed eventually. 
 
But honestly, at this point, I would be happy to know if I'm just an idiot and nothing is wrong with my card other than flying too close to the sun. I honestly probably don't know enough about how monitors and refresh rates draw on GPUs, so maybe this was all just a learning exercise. If something is wrong with my card, however, I'd like to try to RMA it or something.
 
What do y'all think? Help me come down to earth?

 

loyalty4life
Did you get it fixed?

No, I haven't been able to fix it yet. 

Try using ddu and when u go install new driver do a custom clean install. See if that helps

Tried the DDU uninstall and revert to old drivers. No luck. I am really starting to feel like this is temp related. The GPU turns off at 77-78 degrees on three separate tests. As far as I knew, it never did that before? Maybe I just never reached that limit. Could trying to reach a higher hz be to blame? Should I be able to run 120hz on high settings 1440p without an issue (thermal throttling would be expected, I understand, but a complete shut off just seems odd?)

Max allowed temp for the 2080 ti is 89c, so 76-77c is fine. Are you running the fans at max speed?

radioactivepal
Hi Sajin, I did not run the fans at max speed yet. I set the px1 to aggressive the fans would go higher quicker. Just to confirm, I believe 76-77 to be an OK temperature normally, but I'm saying it is consistently crashing at that temperature. Also, weirdly, when I turned gsync off and ran a 3dmark test with my monitor 120hz but nvidia control panel at 60hz, it passed the test and was going over 60 FPS the entire time, mostly in the 100s. I'm wondering if it has something to do with gsync? 

I’m thinking there is something wrong with the card. Don’t think g-sync is the issue, but anything is possible. Will need further testing to confirm.

I just know I'm going to come back here tomorrow saying I jinxed it, but I had some positive progress today after re-doing a couple of the troubleshooting steps:
 

• Took out GPU, cleaned radiator with compressed air, wiped contacts with an alcohol wipe, cleaned PCI-e slot and wiped it down with alcohol, made sure to get in slot groove
• Reseated GPU in the 1st motherboard PCI-ie lane (I had swapped it as part of a troubleshooting step, but since it wasn't working, I moved it back)
• Made extra sure the PSU cables were secure in the GPU 8-pin headers
• Loosed the screws keeping the GPU aligned with the PCIe-lane and IO slot -- too tight and it looked like the card was pending upward in the slot
• Probably took about 10 minutes or so; is it possible being out of the PC that long caused something to reset?
• Restarted computer

I don't know WHAT if ANYTHING that little would have done and I basically feel like I did all of that before (minus the looser screws), BUT:
 

• Ran a 3D mark stress test (Fire Strike) -- I kept Gsync on even though it advises against it (monitor at 5120x1440p/120hz): 3D mark says it didn't pass by like .3 points, but honestly to me it PASSED because the dang GPU didn't crash -- during this test the GPU got as hot as 78 degrees -- previously it had crashed in this range
• Ran Time Spy benchmark twice, got a score of 12,000 some (seems reasonable for 5120x1440p 120hz and again BETTER than the GPU crashing)
• Played Crusader Kings III at 5120x1440 @ 120hz for about half an hour. GPU nearly got to temps that caused the GPU to fail, but the game wasn't stressing it that much this time around

I honestly have no idea what could have changed, but it feels like it's working again. I am sure I will be back here tomorrow in tears, but I feel like this is progress. It's superstitious, but I am afraid to like change anything now that I have games and tests running at this resolution and 120hz). Also afraid to update to the latest drivers again. 
 
I'll check in again to say if it's still stable, but thank you to everyone who replied to this thread, especially Sajin.

Well, I jinxed it. Woke up this morning, woke the computer up from suspend. Ran a stress test and I did it without crashing. So I did the test I had to do to be extra sure. I restarted the computer and redid the stress tests and now I have the crashes again. I guess I need to start the RMA process at this point, unless anyone things reflashing the card with the same bios would help (I would need instructions to know how to do it though)