Screen turns dark and fans start spinning. Discord and other sounds remain audible.
Screen turns dark and fans start spinning. Discord and other sounds remain audible.
Hello,
I recently (1-2 months now)found an issue with my PC and i do not know what causes it. I am currently at the stage that i think that my Graphics card is just dying.
To start whenever i play games i get a Blackscreen at random times. Event viewer is not showing anything and there are no other abnormalities.
I have troubleshoot my Rig with 4 people and no one knows what it could be so i am coming here to see if you have any ideas.
So far i tried the Following:
- Replaced the PSU with a new one with more Power (From 700W to 850W)
- i reinstalled Windows
- reinstalled Drivers with DDU
- reset BIOS
- Updated BIOS
- checking for Overheating, which resulted in normal temps and no signs of Overheating.
- Undervolting the GPU to 80% power draw (This one helped a lot actually which is why i got a new PSU since i thought these are signs of the PSU not delivering enough Power so the Graphics card crashes)
So far the Games or Situations it happens in variate between games and conditions. the only common thing is that it is always under load not even heavy ones. The Performance in games are normal so far so no throttling at any point.
Games that the crashes happened in so far:
- GTA 5 Enhanced
- Battlefield 6 (not after the PSU update)
- Returnal
- Steam VR
My Specs:
Edition Windows 11 Pro
Version 25H2 (and 24H2)
Mainboard: TUF Gaming Z490-Plus (WIFI)
Processor: Intel® Core™ i9-10900K CPU @ 3.70GHz (3.70 GHz)
Installed RAM: 32.0 GB (31.9 GB usable)
Graphics Card: Nvidia RTX 3080 TUF Gaming
Graphics Card Driver: Games Ready Driver 581.57
If anyone has any knowledge about what is happening and if i am right that my 3080 is dying then please do let me know because i am at my limit in terms of what i can do or what it is.
Thank you and have a great day.
It felt quite alike to the issues I experienced after upgrading to a 6800xt. The screen would freeze during games, yet the audio remained intact. I looked up my GPU manufacturer's site and found an updated VBIO, which I installed successfully. Afterward, everything ran smoothly.
The specifications of the power supply are listed for review purposes. Hardware models may vary, but two similar units could show the same issues. Under heavy load, the RTX 3080 exhibits noticeable voltage spikes lasting a few milliseconds, indicating transient surges rather than constant demand. The ATX 3.1 documentation states that premium PSUs support up to 200% load transients, though these are more expensive. Purchasing a 1kW PSU to address the issue carries risk, especially for a mid-range GPU, and may not be necessary. Testing the GPU in a better-specified system provides the clearest evidence. It seems many users rely on word-of-mouth rather than technical data about power supplies. The question of whether to expose a system with a questionable GPU is risky—would you risk your setup for the sake of others?
I upgraded to a beQuiet 850 Watt PSU. Afterwards everything worked almost like before, but the system started having problems again after installation. However, all the games I mentioned before ran smoothly without any issues. A friend of mine has a 700 watt PSU with a 3080 and doesn’t experience those problems, so honestly I’m not sure. I can’t find a 1k PSU.
I tried that before but i can check again after my recent PSU upgrade will update later
If there's some easy tries like disable Variable refresh rate/gysnc, give it a whirl (nvidia settings)
Monitor VRAM Temps: Download GPU-Z (free). Run a stress test like FurMark or play a game, and watch the memory junction temp. If it hits 95–110°C, that's your smoking gun—thermal throttling kicks in, causing the black screen
Already tried vbios update? Huh not sure how I would feel about that - new gpu and I have to update the bios? Never before. That's a welcome to the wacky wurld.
If in warranty, rma card. If there's nothing wrong with it, they send it back. Only reasonable step since messing with another system isn't risk free.
Being a bit off rec. specs might not be terrible if the be quiet is a tier A model. Could still be one of a variety of less than tier A units though. Exactitude of details really is the difference between knowing what it is or not.
Many people confuse temperatures around 75-80°C as normal since they seem low, but if the fan speed exceeds 80-85% under load and the hotspot is more than 15°C above the GPU temperature, the GPU might actually be overheating and require replacement. I frequently encounter this issue when assisting users with their 30 series GPUs. The 850-watt power supply should comfortably handle the system.
I played Battlefield 6 once more for a single round. It was unexpectedly free of black screen. Based on what I observed during the session, here are the key temperatures:
GPU Temperature [°C]: 61 degrees,
Hot Spot [°C]: 82 to 85 degrees,
Memory Temperature [°C]: 82 degrees.
All readings were at their maximum under load and matched the highest values recorded while logging this component.
If my interpretation is accurate, it could indicate overheating. I’d appreciate confirmation if that’s what you were referring to when mentioning overheating.
If needed, I can share the full log for your review.
The differences lie in the hotspot temperatures and fan speed settings. A higher fan speed when under heavy use usually indicates the card is managing increased or higher temperatures, helping it stay within acceptable limits. This also varies with the GPU load, as lighter tasks let the card run cooler due to reduced power consumption. When properly recorded, verify the fan speed RPM and percentage at both low and high temperatures for the GPU and hotspot.