Search results for query: nvidia

It is possible to set up multiseat with separate nvidia cards - but probably not using one card with several monitor outputs. Each instance of Xorg (and the nvidia driver) needs to be told which graphics card to use. I did this a long while ago. I would guess it's quite a bit easier now with...

FWFW the NVidia drivers do not support multi-seat no matter what. What should work relatively easily is have 2 different X11 servers on two different VTs. You can then switch with the usual vt switch keystrokes to select one or the other.

...-> suffocate pipeline) Another one is lack of HW pass-through capability (can't pass-through just one device from one PCI) and the notorious nVidia GPU (it is not easy on Docker Linux either). Both of these are a feature, not a bug. Docker (or Podman) run as root, vulnerable libraries in...

cross: I am using that setup: FreeBSD server with Jellyfin service Nvidia Shield Pro with Jellyfin client This works perfectly for me. However, since NFS on Android (at least on the the Shield) is iffy, the server runs both NFS and Samba, the Shield connects via Samba.

..."do I want to use Jellyfin as a streaming server on my new FreeBSD server?" this thread has me concerned that it might not be wise. I have an nVidia Shield TV Pro, running Kodi, and I have for years been using NFS. But, even more-so now with Android (had LibraElec on a chromebox before)...

If the issue is happening when you're using picom, nvidia is working on it but even the latest Production branch of driver 580.126.18 states it fixed. Known workaround for it is to disable picom. I've never used picom, but using compiz in conjunction with Mate DE on RTX A400 (Ampere generation...

List for 580 and before here. nvidia regularly provides the list per driver releases in READMEs. Note that GPUs supported by both newer and older versions are dropped from the list for older. So if you want full list for 470 series, you need to look into the one for 470.

...follows mouse so I usually have to tweak things so windows don't autoraise on me). I'm at the point of "out of ideas" other than "there is a nvidia-drm-515" plus a few other versions. I'm not sure what the difference is between the nvidia-drm-61-kmod and nvidia-drm-66-kmod but may be worth a...

I'm using Xfce, and the settings are all out-of-the-box. It's set to click to focus, and no changes are happening between the NVIDIA driver switching when I change versions from 580 to 470. Here is the output of xfconf-query -c xfwm4 -lv in case it helps: /general/activate_action...

Yes, the 1650 GPU is supported with the latest 580 driver. https://www.nvidia.com/en-us/drivers/details/264091/ The 470 driver should work as well. It's listed as a legacy driver. https://www.nvidia.com/en-us/drivers/details/226762/

If that’s the case then why has it only happened since I updated the NVIDIA drivers? And why do the issues disappear if I install a previous version of those drivers?

Don't think that is an nvidia thing at all. It looks like some problem with raising and lowering windows in your DE.

...related to this post, but I'm afraid I don't know enough about the subject to be sure. I am wondering if someone could help me. I am using a NVIDIA GeForce GTX 1650 on FreeBSD 15-RELEASE, running Xorg (not Wayland yet). I usually keep my graphics driver packages locked as, in the past, I...

A lot of supercomputers use Modern Fortran, including the PGI CUDA Fortran compiler (now part of the NVIDIA HPC SDK), that lets you do parallel programming on an NVIDIA CUDA GPU. Although CUDA support on FreeBSD is apparently getting better, it still does not support parallel programming with...

Well, I tested - without logout - but no success for me. Do you think logout/login would change something...?

Just to be sure, does the situation changes if you flip Sync to VBlank with x11/nvidia-settings? Wouldn't help for flickering, but could make differences in tearing.

.... # Definition of LINUXKPI_VERSION # # This is used to track linuxkpi compatability with # the source tree and external components such as # nvidia-drm LINUXKPI_VERSION_NUMBER= 61000 Considering how long it takes to port linuxkpi (6.9 was released Aug 9, 2025, and 6.10 on Jan 29. 2026)...

I found it, it was that one : https://bbs.archlinux.org/viewtopic.php?id=311088 Their answers are to disable compositor or reverse to a very old driver version - which is not a goog idea... Strange thing, one of the last messages of the thread says it happens on AMD.

...was not provided by any .service files") Bottom portion of the console's output, after choosing Project from the Splash Screen OpenGL vendor "NVIDIA Corporation" OpenGL renderer "NVIDIA GeForce GTX 960/PCIe/SSE2" OpenGL threaded? true OpenGL ES? false OpenGL maximum texture size = 16384...

Also, Mate DE + Compiz on FreeBSD is stable, too. Not all of Compton based compositors are affected.