F5F Stay Refreshed Hardware Desktop Problem with minimal GPU activity on RTX 3080

Problem with minimal GPU activity on RTX 3080

Problem with minimal GPU activity on RTX 3080

R
Rounyx
Posting Freak
838
11-14-2021, 02:19 PM
#1
I've been facing a rather odd problem lately, and I'm hoping someone can help me out. Over the past few days, my GPU wasn't being used to its full potential—I wasn't too concerned at first, thinking it might be related to the game I play (Dying Light, which had a memory leak). But recently, in games like Black Ops 7 Beta and Atomic Heart, the usage dropped to about 60–70%, and my FPS plummeted significantly.

I tried using DDU in safe mode and reinstalled the latest NVIDIA driver with NVCleanstall, which fixed the issue temporarily—only for that specific night. Now it's happening again. What's unusual is that everything works perfectly during the day: full GPU usage, normal temperatures, and high FPS. But in the evening and at night, the GPU suddenly underperforms, drawing less power (now around 190W instead of 300W+) and overall performance drops.

Currently, I'm playing BF6 with 110–95 FPS at 2K, high settings, and DLSS Q. However, later in the evening it drops to 40–50 FPS. My GPU is over 5 years old, and I haven't replaced the thermal paste or pads. Could it be that the old paste/pads aren't transferring heat properly, causing throttling? Or could something else—like my PSU, drivers, or power limits—be responsible?

I asked ChatGPT (lol), and they said my GPU is running at a very high hotspot (99–105°C during gaming). According to ChrisTitusTech, high performance mode in Windows Power Plan also shows this. DDU cleanup and driver reinstall didn't help. I'm leaning toward maximizing performance via the NVIDIA Control Panel. It almost feels like there might be a power limit, voltage issue, or even system-level throttling during the evening. Or maybe something else is at play?

My specs are:
- GPU: Gainward GeForce RTX 3080 Phoenix
- CPU: Ryzen 9 5900X
- RAM: 32GB @ 3200Hz
- Motherboard: ASUS Prime X470-Pro (last BIOS)
- OS: Windows 11 Pro
- NVIDIA Driver: 581.42
- PSU: SeaSonic PRIME Ultra 750W 80+ Titanium (SSR-750TR)

Any advice would be greatly appreciated.
R
Rounyx
11-14-2021, 02:19 PM #1

I've been facing a rather odd problem lately, and I'm hoping someone can help me out. Over the past few days, my GPU wasn't being used to its full potential—I wasn't too concerned at first, thinking it might be related to the game I play (Dying Light, which had a memory leak). But recently, in games like Black Ops 7 Beta and Atomic Heart, the usage dropped to about 60–70%, and my FPS plummeted significantly.

I tried using DDU in safe mode and reinstalled the latest NVIDIA driver with NVCleanstall, which fixed the issue temporarily—only for that specific night. Now it's happening again. What's unusual is that everything works perfectly during the day: full GPU usage, normal temperatures, and high FPS. But in the evening and at night, the GPU suddenly underperforms, drawing less power (now around 190W instead of 300W+) and overall performance drops.

Currently, I'm playing BF6 with 110–95 FPS at 2K, high settings, and DLSS Q. However, later in the evening it drops to 40–50 FPS. My GPU is over 5 years old, and I haven't replaced the thermal paste or pads. Could it be that the old paste/pads aren't transferring heat properly, causing throttling? Or could something else—like my PSU, drivers, or power limits—be responsible?

I asked ChatGPT (lol), and they said my GPU is running at a very high hotspot (99–105°C during gaming). According to ChrisTitusTech, high performance mode in Windows Power Plan also shows this. DDU cleanup and driver reinstall didn't help. I'm leaning toward maximizing performance via the NVIDIA Control Panel. It almost feels like there might be a power limit, voltage issue, or even system-level throttling during the evening. Or maybe something else is at play?

My specs are:
- GPU: Gainward GeForce RTX 3080 Phoenix
- CPU: Ryzen 9 5900X
- RAM: 32GB @ 3200Hz
- Motherboard: ASUS Prime X470-Pro (last BIOS)
- OS: Windows 11 Pro
- NVIDIA Driver: 581.42
- PSU: SeaSonic PRIME Ultra 750W 80+ Titanium (SSR-750TR)

Any advice would be greatly appreciated.

B
bdog4444
Junior Member
9
11-15-2021, 05:19 PM
#2
PSU: SeaSonic PRIME Ultra 750W 80+ Titanium (SSR-750TR)
What is the age of your PSU?
What is the make and model of your motherboard? What BIOS version are you using?
What is the make and model of your case?
The GPU has been running for over five years, and I haven’t changed the thermal paste or pads. Is it possible that the old materials aren’t transferring heat effectively, leading to throttling? Or could another factor—such as the PSU, drivers, or power limits—be responsible?
It might help to perform a thorough cleaning of the GPU and see if that improves performance. You could also consider replacing the thermal pads and paste with a higher-quality alternative.
B
bdog4444
11-15-2021, 05:19 PM #2

PSU: SeaSonic PRIME Ultra 750W 80+ Titanium (SSR-750TR)
What is the age of your PSU?
What is the make and model of your motherboard? What BIOS version are you using?
What is the make and model of your case?
The GPU has been running for over five years, and I haven’t changed the thermal paste or pads. Is it possible that the old materials aren’t transferring heat effectively, leading to throttling? Or could another factor—such as the PSU, drivers, or power limits—be responsible?
It might help to perform a thorough cleaning of the GPU and see if that improves performance. You could also consider replacing the thermal pads and paste with a higher-quality alternative.

S
sorensej
Junior Member
9
11-16-2021, 05:05 PM
#3
Hey there, thanks for the update!
PSU info: from December 2018.
Mobo: ASUS Prime X470-Pro with BIOS 6232.
Case: NZXT H6 Flow – I’ve got a custom dust filter for it.
GPU: I’m hoping it works out well; I don’t want to spend time or money on something that might not be worth it.
S
sorensej
11-16-2021, 05:05 PM #3

Hey there, thanks for the update!
PSU info: from December 2018.
Mobo: ASUS Prime X470-Pro with BIOS 6232.
Case: NZXT H6 Flow – I’ve got a custom dust filter for it.
GPU: I’m hoping it works out well; I don’t want to spend time or money on something that might not be worth it.

J
JenpaiMC
Member
106
11-22-2021, 06:34 AM
#4
Hey there, thanks for the update! Just wanted to assist with the details.
J
JenpaiMC
11-22-2021, 06:34 AM #4

Hey there, thanks for the update! Just wanted to assist with the details.

W
Will2640
Junior Member
8
11-22-2021, 01:48 PM
#5
Hey there, just wanted to share a quick update. I finally figured out the problem—it was all about the GPU hotspot. After swapping out the thermal paste and pads on my RTX 3080 with a pro, everything worked perfectly again! The temperatures are back normal, power consumption is full, and FPS is stable, almost like before. It’s really impressive how much of an impact this made. If your GPU is around 3–5 years old, I strongly suggest replacing the thermal paste and pads—it’s definitely worth it. Everything’s now working perfectly! Thanks for the help, mate.
W
Will2640
11-22-2021, 01:48 PM #5

Hey there, just wanted to share a quick update. I finally figured out the problem—it was all about the GPU hotspot. After swapping out the thermal paste and pads on my RTX 3080 with a pro, everything worked perfectly again! The temperatures are back normal, power consumption is full, and FPS is stable, almost like before. It’s really impressive how much of an impact this made. If your GPU is around 3–5 years old, I strongly suggest replacing the thermal paste and pads—it’s definitely worth it. Everything’s now working perfectly! Thanks for the help, mate.