RTX 2080 Ti black screen

[acegard]

This problem started happening today. After about 3 minutes of stress on my watercooled Aorus Xtreme 2080 Ti (Whether it's FurMark or games), both of my monitors go black. I can still hear audio but there is no response from the KB/mouse and only a hard reboot recovers. Digging into event Viewer reveals occasionally "Display driver nvlddmkm has stopped responding and has recovered." but sometimes does not appear to trigger this error. 

I have tried: Googling fixes for the above Event Viewer code. I have underclocked the GPU. I have tried disabling XMP, disabling G Sync, trying different monitors and different peripherals. I have tried rolling back the drivers to a point which I remember them being stable (last week) although it does not seem to correlate with a driver update. Temperatures are fine, only hit about 66 degrees by the time the error happens. This does not happen under "casual" use. Yesterday, it was working fine! I have been able to play Destiny on it though this evening when my friends and I launched Sea of Thieves the problem manifested and hasn't gone away. I could really use some help here, thanks. 

 

GPU: RTX 2080 Ti Aorus Xtreme no add'l OC

CPU: Ryzen 3800x no OC

Motherboard: Asus X470-F Gaming

RAM: 64GB TridentZ RGB NEO 3600Mhz, 4x16 XMP enabled

PSU: Corsair RM850x

Operating System & Version: Windows 10, latest updates

GPU Drivers: 442.74 (Rolled back)

[Tostr4]

I had the same issue a few years back. In my case it was my extra PSU who failed to maintain a high enough output to power the fans of mij GPU. To protect the GPU, it shuts down completely, without warning. 

[Shepherd_Strike]

28 minutes ago, acegard said:

This problem started happening today. After about 3 minutes of stress on my watercooled Aorus Xtreme 2080 Ti (Whether it's FurMark or games), both of my monitors go black. I can still hear audio but there is no response from the KB/mouse and only a hard reboot recovers. Digging into event Viewer reveals occasionally "Display driver nvlddmkm has stopped responding and has recovered." but sometimes does not appear to trigger this error. 

I have tried: Googling fixes for the above Event Viewer code. I have underclocked the GPU. I have tried disabling XMP, disabling G Sync, trying different monitors and different peripherals. I have tried rolling back the drivers to a point which I remember them being stable (last week) although it does not seem to correlate with a driver update. Temperatures are fine, only hit about 66 degrees by the time the error happens. This does not happen under "casual" use. Yesterday, it was working fine! I have been able to play Destiny on it though this evening when my friends and I launched Sea of Thieves the problem manifested and hasn't gone away. I could really use some help here, thanks. 

 

GPU: RTX 2080 Ti Aorus Xtreme no add'l OC

CPU: Ryzen 3800x no OC

Motherboard: Asus X470-F Gaming

RAM: 64GB TridentZ RGB NEO 3600Mhz, 4x16 XMP enabled

PSU: Corsair RM850x

Operating System & Version: Windows 10, latest updates

GPU Drivers: 442.74 (Rolled back)

Your gpu might be faulty thats what happened to my 1070

[acegard]

10 hours ago, Tostr4 said:

I had the same issue a few years back. In my case it was my extra PSU who failed to maintain a high enough output to power the fans of mij GPU. To protect the GPU, it shuts down completely, without warning. 

Good idea, but there were no issues with the PSU before yesterday. It is a relatively new unit. 

[MANDINGO]

Maybe your hdmi or displayport?

[acegard]

For posterity, I fixed it! It turned out to be a bad mount of the waterblock. The thermal paste was not covering the die evenly so, even though the overall temperature was indicating fine, there were hot spots when the card powered up for gaming and it was crashing. So, I manually spread the paste while remounting my Phanteks block and now it works fine. Temperatures in the 50s after a healthy overclock under load. Thanks for all your help!