Issue: Whenever I go to play a game like Warzone, Cold War, ESO, Tarkov within 1-6 minutes of entering a game/match the screens freeze, go black, audio stutters then stops... PC stays on, fan lights set back to default and keyboard is unresponsive. The only way taction I am able to do is a hard reset with the power button on the case.  

Event Viewer is should Kernel 41 power, so no real logs.. On the off chance I get a memory dump saved It points towards: 

“VIDEO_TDR_FAILURE (116) Attempt to reset the display driver and recover from timeout failed. Arguments: Arg1: ffffc18b1b0a1460, Optional pointer to internal TDR recovery context (TDR_RECOVERY_CONTEXT). Arg2: fffff801700019e4, The pointer into responsible device driver module (e.g. owner tag). Arg3: ffffffffc000009a, Optional error code (NTSTATUS) of the last failed operation. Arg4: 0000000000000004, Optional internal context dependent data.  Unable to load image \SystemRoot\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_4a746d937e6a7240\nvlddmkm.sys, Win32 error 0n2 *** WARNING: Unable to verify timestamp for nvlddmkm.sys *** WARNING: Unable to verify checksum for win32k.sys” 

Figured it was the Power in the beginning, RMA’d my 3 year old EVGA SuperNova 1000w... Received a brand new unit, placed it in my system with all new power cables and same issue.  

Then I brought it to micro center to confirm it was the card, they verified....  

Submitted the RMA, 2 weeks later I have my “new” card and same issue. I can tell this thing is just refurbished due to ware on it.. Small scratches here and there from what appears normal use.  This card is actually in worse shape than the one I originally bought. It crashes way quicker.  

 

In the mean time I’ve been using my EVGA GeForce GTX 1080 ti FTW3 with no issues whatsoever.  

 

I’ve tried: 

Enabled Resize Bar 
Modified Chrome 2D settings 
Updated Bios and Vbios 
Rolled back drivers to v456 via safemode and DDU 
Disabled Asus Aura 
Disabled iCue 
Disabled EVGA Precision X 
Edited Registry to allow for a longer delay time 
Adjusted Nvidia Control Panel Settings (max power, gsync off, etc.) 
Tried Afterburners and modified settings there 
Disabled XMP Profile for memory 
Underclocked by 150 
Set frame targets, 60 works for the longest time before crash, about 6 minutes 
G-sync on and off 
Monitor fixed and variable refresh rates 
Setting fixed game fps 
Checked hard drives health and they are in good shape, all Samsung ssd’s and m.2.  
Ran a ram diag via windows client on startup  

Temps look to be in line, GPU hotspot gets up to almost 80* some times but overall never gets into the 70's. 
 

Running out of ideas here, any suggestions or help is appreciated.  

 

Specs in image  , if you can't see
Intel Core i7-11700k
Asus Prime Z490-A bios v2103
32gb (4x8gb) Corsair Vengeance RGB Pro 3200
EVGA 3090 FTW3 Ultra (1080ti in here right now)
 
 

The bug check is explained here in full details:
Bug Check 0x116 VIDEO_TDR_FAILURE - Windows drivers | Microsoft Docs
 
I would verify if your case is providing enough air flow, the various GPU temps can spike up very quickly depending on the workload, faster than the software could register.
Try a home fan that blows directly on the GPU and see if you get more play time before it dumps again.
Otherwise Sajin may be right, you may have received another dud.
 

I have been having the same exact problem for months in warzone. I get a few games in and then a complete freeze. I've tried absolutely everything.  Wondering if it's a bad memory module or something?  

the only temporary fix i have at the moment is to reset the application every few games.  But mine is not freezing as quickly as yours is i dont think. Have you reached out to EVGA support yet?  I just submitted a ticket. 

So I've ran a memtest 86 and my ram passed..
 
I need to test more but I think I have it working.


What I did that fixed it: In the order that I did it, may not be the optimal or correct order
-Edited registry values to include delay on TDR failures - 
-G-sync off
-No EVGA Precision x1 software
-Made sure extra airflow on card, fan curve running off GPU temps... Open sides on the case
-Wiped both ssd's that I had my games on and reinstalled them all
-Moved the GPU into PCIe slot #2 (booted, put the card in debug mode and tried to play games... successful) moved the gpu back to PCIe slot #1 with Gen 4 x8 lanes enabled in bios
 
I played a bunch of cold war and warzone lastnight without crashes (Previously I would crash within a minute of starting a match)... I could tell where the tdr timed out and stuttered before it restart... need to do ESO, BF4 and Tarkov today to test further. I also watched temps and power in HW Monitor the whole time. So far so good, fingers crossed it is a fix for me. Hope this helps someone else and I'll update in a few days of more test.