Question Black screen after 5-20 minutes in GPU intensive games

[Anom1c]

I have a pretty old PC but it always worked fine for 1440p60 gaming until now. I don't know what caused it because I only play intensive games sporadically. Recently I re-applied cooling paste to both the GPU and CPU because it had been years ago since I last did it. Could that have caused this?

I wanted to to play some GTA V first and after 5-15 minutes got weird lines + crash and error "d3d init error", tried reinstalling directx and that didn't fix it.
So I gave up on that game and played AoE2😀E without problems for days and hours at a time.
Now I'm trying to play Kingdom Come Deliverance and after 5-15 minutes I get a black screen but the game doesn't crash and I don't get any error message. I can even ALT TAB out and back in the game and get a moveable game cursor and game sound for a fraction of a second.

So I googled possible solutions and tried these with no succes:
- Re-install DirectX
- Update Bios
- Remove and Reinstall GPU drivers in safe mode with DDU
- Launch game as Admin & W7 Compatibility mode
- Turn off FPS counter
- Checked inside GPU for paste spillage / damage - see pic

Spoiler: Pic

https://imgur.com/a/mxHhn7c

View: https://imgur.com/a/mxHhn7c

(I cleaned it before re-applying, the mess was either already there or from the previous time I re-applied paste, I can't remember if I ever did it before)
- Checked GPU slot for dust and cleaned the GPU
- Made log with GPU-Z and did Furmark Test - Max Temp is 77 degrees Celsius, hotspot 91
- Changed GPU Fan Curve to 100% starting at 70 degrees

Only thing within my power that I haven't done yet is a fresh OS install but I really hate doing that lol

Any ideas?

Specs:

Spoiler: PCP

PCPartPicker Part List

CPU: Intel Core i7-8700K 3.7 GHz 6-Core Processor ($169.95 @ Amazon)
CPU Cooler: CRYORIG R1 Universal 76 CFM CPU Cooler ($0.00)
Thermal Compound: Arctic Silver 5 High-Density Polysynthetic Silver 12 g Thermal Paste ($19.98 @ Amazon)
Motherboard: Asus Prime Z370-A ATX LGA1151 Motherboard ($316.49 @ Amazon)
Memory: Corsair Vengeance LPX 32 GB (4 x 8 GB) DDR4-2133 CL15 Memory ($0.00)
Storage: Samsung 950 PRO 256 GB M.2-2280 PCIe 3.0 X4 NVME Solid State Drive ($0.00)
Storage: Western Digital Red 8 TB 3.5" 5400 RPM Internal Hard Drive ($259.99 @ Amazon)
Storage: Western Digital WD_BLACK 1 TB 3.5" 7200 RPM Internal Hard Drive ($0.00)
Video Card: EVGA Superclocked+ ACX 2.0+ GeForce GTX 980 Ti 6 GB Video Card ($0.00)
Case: Fractal Design Define R5 ATX Mid Tower Case ($0.00)
Power Supply: Corsair AX760 760 W 80+ Platinum Certified Fully Modular ATX Power Supply ($0.00)
Optical Drive: Samsung SH-224FB/RSMS DVD/CD Writer ($0.00)
Operating System: Microsoft Windows 10 Home Retail - Download 32/64-bit ($0.00)
Monitor: BenQ BL3200PT 32.0" 2560 x 1440 60 Hz Monitor ($0.00)
Keyboard: Corsair K70 RGB MK.2 Wired Gaming Keyboard
Mouse: Logitech G502 Proteus Spectrum Wired Optical Mouse ($69.98 @ Amazon)
Headphones: Beyerdynamic DT 990 Pro 250 Headphones ($0.00)
Total: $836.39
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2024-12-14 11:13 EST-0500

 

[Lutfij]

How old is the PSU in your build? You should source(borrow, not buy) a reliably built PSU that can deliver 650W of power to the entire system and see if the issue recurs.

Just an FYI, you've left performance on the table by working with DDR4-2133MHz rams. You'd see an uplift in your system's performance working with a 2x16GB DDR4-3200MHz, tight latency ram kit.

 

[Anom1c]

Everything was bought late 2015. Could it be the PSU? It was one of the best at the time and I saw that my GPU under 100% load draws only 250watt.

At the time I bought it it had a 5820k which broke along with the mobo so I just switched to 8700k. The RAM is actually Cas 13 and was the sweet spot for that CPU iirc.

 

[Anom1c]

Could this be thermal throttling? FPS stays around 60 though

Spoiler: Pic

https://imgur.com/sUTxAsP

View: https://i.imgur.com/sUTxAsP.png

 

[stonecarver]

If by doing what Lutfij suggested by verifying the issue with a loaner power supply.

Also are the PCI-E power cables firmly plugged into the GPU?

If problem solved no need to keep reading.

If it still has the same issue I would consider taking the GPU apart again and just reassembling. Could be something just slightly offset. It happens.

 

[Anom1c]

I re-installed windows and that didn´t fix it, I could play for longer as usual though. Just when I was pretty sure it was fixed, it happened again...

So I borrowed an old GTX1060 and it hasn't happened anymore (for hours).

So there is a hardware problem with my 980ti that I have no idea what to do about...

 

[SkyNetRising]

Any ideas?
Thermal Compound: Arctic Silver 5 High-Density Polysynthetic Silver 12 g Thermal Paste ($19.98 @ Amazon)

You have to be careful with thermal pastes containing silver.
While it is not exactly electrically conductive, it is still slightly electrically capacitive.

On your photo you have smd elements drenched in thermal paste.
This is a big no-no with silver containing pastes.

You have to carefully clean that off and apply thermal paste on gpu crystal only.

 

[ohio_buckeye]

The good thing is as Will you can upgrade to something even as cheap as an rx 6600 for under 200 dollars and it is about the performance of a gtx 1080 iirc.

 

[Anom1c]

Yeah I think I did re-apply thermal paste once, but way too much... Was planning to buy a whole new build anyway in 2025. I didnt use the arctic silver for the GPU though, I had another one that came with the cryorig cooler that I dont think is conductive.

For now I guess I'll keep using that 1060... maybe mess around and try to clean the 980ti a bit more