I have a Nvidia GPU and proprietary prime drivers since I have an Intel CPU with an iGPU, but the iGPU is never used I think.
I don’t have the “prime-select” command, I can’t use it, it says bash: prime-select: command not found, and I’ve never seen any secondary GPU in configurations of the system, also I can’t see anything if I connect a secondary monitor to the motherboard HDMI. It’s like the iGPU is not there at all (it’s enabled in BIOS).
I have several questions.
I used the Manjaro hardware tool to install this drivers, so idk why is this happening or what should I do.
Also, I’ve seen a lot of problems regarding this (I mean using both GPUs).
-Is it worth to even enable the iGPU at this point? (I’m on a desktop pc, not laptop) or will I have more problems than advantages?.
-What would I win using it? more performance? since some things will be loading on iGPU and the Nvidia GPU will be focused on heavy things?
-And will those “optimizations” happen easily or will I have to be coding and managing hard configurations to make the desktop render with iGPU and games with Nvidia for example?.
-And what do I need to do to use it? installing prime drivers is not enough?.
Please note that I’m not a professional programmer or Linux expert when you answer.
Yes, I can see “00:02.0 VGA compatible controller: Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor Integrated Graphics Controller (rev 06) (prog-if 00 [VGA controller])” in your command output.
About the 3rd point, the thing is that all games I run are for sure using Nvidia GPU, because the performance is good and I see the GPU usage on Nvidia X Server Settings, also the desktop compositor using Opengl is also on Nvidia, so basically even if the iGPU is active, the whole system is only using Nvidia’s dGPU.
Edit: I tried disabling the dGPU on Lutris and even then, games always use dGPU for running. Also on Yuzu or Citra, both of them are using dGPU, so it’s not a Lutris/Wine thing.
Also, what about the second monitor? why I can’t use it when connected to the motherboard?.
Ok… I am bit confused why mhwd did not create a config at /etc/X11/xorg.conf.d/90-mhwd.conf. Anyway… The problem here is that xorg is instructed to use the nvidia gpu only at /etc/X11/xorg.conf. Remove this file and reboot… then the Intel driver will configured first automatically. Nvidia needs always a config file and therefore xorg will not touch it.
But do you think it’s worth? I mean, my system is not unstable or something, it works fine, I wonder if there is any sense in using the iGPU, and having to specify every time I want to run something with dGPU.
Maybe, maybe not… If you use your desktop computer normally without any tinkering and you use the nvidia screen outputs, then its not worth to touch it. I mean I have a Radeon iGPU on my Desktop and the only thing I do with it is just passthrough it to VMs for better performance (not for gaming).
I don’t need VMs so much, so it’s probably not worth for me then.
Btw, in case that one day I want to do it, I noticed that there is a file called 90-mhwd.conf in /etc/X11/xorg.conf.d/ that points to an empty file called nvidia.conf in /etc/X11/mhwd.d/. Is that a problem? if I just delete the file you said, it will work or do I need to delete this two files too?
But then if I delete /etc/X11/xorg.conf, the second one will be used, but since it’s an empty configuration file, there will be no configuration nor it will create a new xorg.conf file right?. Or did I misunderstood something?