PC shuts down and the cooling fans start making noise?
PC shuts down and the cooling fans start making noise?
Hello,
This issue began about a month ago and has become more frequent. While playing games, my monitor sometimes turns completely black and the GPU fans become very loud. I can still hear audio, but during Discord calls people don’t pick me up, and the only solution is to restart the system. Up to now, I’ve checked temperatures and run stress tests without any crashes. It’s happening a few times each night, and I’m unsure what to do next.
Specs:
CPU – Intel Core i9-23900KF
GPU – NVIDIA GeForce RTX 3080 10GB
RAM – Corsair Dominator 32GB (16GB x 2) DDR5 5600MHz
PSU – 1000 Watt NZXT C1000 80 PLUS Gold
MOBO – ASUS PRIME Z690-P
I measured the GPU hotspot during the last crash and found it was about 20°C higher than the GPU temperature, but it was only around 95°C at the time. It’s unclear if that contributed to the problem.
The GPU is dust-free? If you blow it out, don't let the fans spin too much!
Check the side panel and remove it. For testing, use a fan to blow air into the PC.
Continue using your usual computer habits and play games that cause crashes.
We'll see later if it's still crashing during games or if the fan helps prevent it.
No crash means heat issue remains.
If it keeps crashing, we'll investigate further.
My GPU wasn't really dusty, so I didn't have to worry about it. I removed the glass and used two small handheld fans aimed directly at the GPU, which allowed me to play without problems last night. However, I crashed within five minutes after starting a game this morning.
This suggests your GPU temperature stays at only +75°C, which is under the +80°C threshold advised on this site (at the end).
They note throttling occurs between 88 to 89°C, indicating your GPU might run around 108-109°C.
The RTX 2080’s thermal limit is near 88-89°C, after which it may throttle and affect performance.
I’ve tested my RTX 3060 at full power (170W) for extended periods during video editing; it remained stable only after a bug fix in Topaz Video AI.
Before that, when Topaz Video AI was unstable, I set the GPU power cap to 95% in MSI Afterburner to prevent system freezes.
I recommend applying a 95% power limit on your 2080 to help identify the issue. It’s not perfect, but it should aid your troubleshooting. Sacrificing a few frames is better than risking instability.
I mistakenly used 2080 instead of 3080, but it's good to know. My current GPU temperatures are around 75°C or lower before it crashes. If this is still a problem with the 3080, I'll look into ways to limit its power usage.
Temperature serves as a sign that the GPU is under strain, yet +75°C isn't sufficient to trigger freezes or reboots. I'd lean more toward software issues. My experience shows that my system with an RTX 3060 frequently crashed in Topaz Video AI during early 2023, until the developers patched the problems. Likewise, game developers sometimes release updates before fully fixing bugs. It's often cheaper to let the end user handle Quality Assurance.
Given your setup of a 3080, I suggest checking the NVidia website for the latest gaming drivers. I typically use their Studio Driver for stability, though it might not suit your needs perfectly.
Once drivers are updated, inspect the PCIe power connections to your card and ensure they're securely connected. Also verify the power supply unit's end. Even with 12VHPWR issues, some older 8-way connectors can overheat occasionally.
If the issue continues across all games, consider installing a budget SATA SSD and reinstalling Windows (after disconnecting other drives). It's possible something has failed in your operating system or applications.
Then, try swapping out another GPU, PSU, CPU, and motherboard. Since this isn't a feasible option for many, do you or a friend have a similar PC that could host the RTX 3080?