High GPU Temperature

Hello all!

First off, i’m a complete Linux noob, so bear with me. Just started using it a little over 1 month ago.
Since i plan to ditch Windows, i have been testing Linux. Started with Ubuntu(didn’t like), Fedora KDE(pretty good), and now OpenSuse Leap KDE(X11).
For the moment i’m dual booting with Windows, having OpenSuse installed on a secondary hard drive, been testing everything out, everything is going fine, got my Jellyfin server up and running.

The problem being, my GPU has higher temperatures on OpenSuse for some reason, for instance, on windows just by having Firefox open, one tab(with this forum), temps are at around 40º, on OpenSuse it reaches 49º, 50º, 51º. What is the problem here? Any setting i need to change. I’ve been trying to minimize this situation by using CoreCtrl. My GPU fans usually only turn on by default at around 50º, so i use CoreCtrl to force the fans on.

Windows

OpenSuse
Screenshot_20250410_150456

My computer Specs:

I5 4460
16GB Ram
RX570 4GB

Thanks in advance.

50C is well within operating specs, so stop worrying about it.

Also, stop directly comparing tools in one OS with different tools in another. HWMon and lm-sensors may not be measuring the same sensor.

And if you are dead set on comparing the two tools, take a closer look at all the data. lm-sensors is also showing that your current power draw is basically twice what HWMon is showing. Almost like it’s a completely different system running a completely different driver.

Possibly drivers were not properly installed.

Post

inxi -aFz

I feel like something is not right, i mean the temperature doesn’t stabilize, even just by using Firefox, on Windows gets to 39º and low 40s and doesn’t increase more, it becomes stable. On OpenSuse it slowly starts increasing till 50º, 51º, and that’s when the GPU fans turn on by default, so the temperatures don’t go any higher.

inxi -aFz

@Slum_Shpongle Well it’s also subjective, how have you confirmed that windows is correct, how have you confirmed linux sensors is correct? Is the GPU sensor a thermistor or diode? Where is the sensor located on the GPU?

Some sensors need calibration with a config file, have you looked at how the GPU fan can be controlled (trigger points) etc?

I totally get your point, don’t get me wrong, just trying to make sure everything is working correctly.

Everything is running within operational specs. Why are you insistent there is something wrong?

50c for a GPU is not hot. If you want it lower, change the fan curve, but 50C is not hot, not unusual and not outside operational specs.

1 Like

Based on what you write there seems to be a real difference between Linux en Windows. I agree with the others that it is not something to worry over but if you want to find the root cause this might guide you in the right direction:

https://bbs.archlinux.org/viewtopic.php?id=244257

I feel like this is exactly the problem, the power draw is higher than usual, hell, even my electric bill was higher this month :sweat_smile: and I’m not even joking! I’ll have a look at that and try to change some setting.

I really appreciate your help guys, i won’t bother with this anymore, I’ll lower the power of my GPU, almost 100% this will fix the issue.

Post text with code tags (</> button).

Use amdgpu driver instead of modesetting one.
That means: properly install drivers for AMD video card.

Mine:

  Display: x11 server: X.Org v: 1.21.1.11 with: Xwayland v: 24.1.1
    compositor: kwin_x11 driver: X: loaded: amdgpu dri: radeonsi gpu: amdgpu
    display-ID: :0 screens: 1

Just want to share, that this fixed my problem, everything is working correctly now, stable temperature. Really appreciate your help.

echo "low" | sudo tee /sys/class/drm/card0/device/power_dpm_force_performance_level
3 Likes