[SOLVED] ZOTAC 3090 Trinity OC -- screens black out randomly

[DHGamerMR]

Having some issues after upgrading from my MSI GTX 1660 6GB to a new Zotac RTX 3090 Trinity OC 24GB a couple weeks ago. Upgraded my PSU from a Corsair RM750x to an RM1000x to make sure I could handle the extra power needs, and I figured this would be a great step up for my Blender and video editing, but it's been anything but helpful so far.

I'd heard about potential heat issues with some 30xx series GPUs, so I was monitoring it closely the first few day or so and then used Afterburner to adjust the fan curve to ramp up quicker. That's kept the temps in the 50-60 C range during gaming, 3D modeling, or editing video, and as low as 28 C otherwise. Nothing new or demanding gaming-wise either

Unfortunately, since the install I've had an issue where my system randomly has all three of my monitors go black and the fan speed skyrockets audibly. Mostly happens when I'm gaming or working in Blender, but sometimes randomly while temps are in the low 30s and I'm just sending emails for work, so it doesn't appear to be dependent on the temps. Still have audio, so video appears to be the only thing affected. Need to hard reset the PC at the power button at that point to reboot (display driver reload via keyboard shortcut does nothing), and hopefully I'm able to save whatever I'm in the middle of before it happens.

Did a full uninstall and clean reinstall of the latest drivers with DDU. Checked all my connections and reseated them to be on the safe side. Still happening and it shows up as LiveKernelEvent 141 in the Reliability History.

I'm running Win10 Pro 64bit with an Intel i7-8700K/MSI MPG Z390 mobo and 64GB of GSkill Aegis DDR4-3000 RAM. Everything is in an older Corsair Carbide 400R Mid Tower case with 6x 120mm fans for lots of airflow and cool temps throughout. Is there anything else I can try here? Or is this just an RMA situation?

 

[LORDPrometheus]

I could tell you how to fix this but a retired mod has a much better post from an earlier thread try his steps---> https://forums.tomshardware.com/threads/screen-goes-black-with-fans-running-at-full-speed.3761221/

 

[Phaaze88]

Is there anything else I can try here?

Adding to the above post by LORDPrometheus:
Are you running psu cable extensions or a PCIe riser device? If yes, please remove them and test.
If using a UPS, try with the PC plugged directly into the wall outlet.
Are you able to try the gpu in another PC and reproduce the same issues?

Or is this just an RMA situation?

If the above suggestions don't help, that would be my next move.

 

[DHGamerMR]

Thanks! I'll look at all of this tonight after work (or more likely tomorrow after work) and reply back with the results over the weekend.

For the record, no PSU extensions or PCIE riser used. I am using a UPS though, so I'll try working without that if the rest doesn't work.

 

[DHGamerMR]

Carved out some time last night after work and followed every step except for a fresh install of Windows 10. Fired up Blender and started working on a low poly landscape for a video I've been trying to work on for a few weeks.

Ten minutes in, the temps had been hovering around 50C for a few minutes, and then I was staring at three black screens. No joy.

No 2nd PC to use as a test so this experiment has come to an end. Time for me to return this and move on to a different GPU. Thanks for the help though!