Nvidia Drivers 3070 Ti

[Jasmine Lemaitre]

Iworry that while stuffing around I may have stuffed the default graphics driver, as well as the Nvidia ones. So I am really not sure where to go from here. All the guides I see say that ubuntu can find the correct nvidia packages, however it cannot oven see the GPU.

Also, it should be noted I was running version 510 (Installed via the GUI in 3rd party apps) before I borked it, and purged everything. I have now installed 550 as it said it is newer and is compatible with my 3070.

If anyone could help trouble shoot this or knows why linux wont even give me a display even on default driver, please let me know. Or if there is any way to restore defaults or even Nvidia, so I can get a display and not have to do it through ssh.

I have issues managing to make my brand new Asus RTX 3070 Dual to work. I previously had a GTX970 that was working perfectly and the nvidia driver installed.

So today I just swapped the boards and nothing seem to work. I have the latest nvidia driver version which I believe is supposed to support my hardware so I'm not sure what is going on.

Temporarily switching to the LTS kernel totally solved the KDE widget icon problem, but as of today I still cannot run Firefox with hardware acceleration turned on while using linux 5.4.75-1-lts and nvidia-lts 1:455.38-3.

You should probably open another thread for this. Maybe including a xorg/journal log. What can happen here regardless of what's what (with every card and driver including OSS ones) is a race condition in loading the graphics subsystem fully before attempting to start xorg.

I did more troubleshooting and here is my conclusion:

prior to linux 5.9 and nvidia 455.38 I was able to use webrender and hw_acceleration in Firefox, even though nvidia was not officially supported by Mozilla. I had these user prefs set:

I have a new configuration and I have several issues. It's an Optimus laptop with AMD iGPU and NVIDIA dGPU (NVIDIA GeForce RTX 3070). I try to setup a dual boot system. So one of my main problems is that my Lenovo Legion 5 does not detects my monitor after I installed the latest NVIDIA v520 driver from the Ubuntu repo. Before the NVIDIA driver installation the system detected correctly both displays. The external is an LG 24GM79G-B and it's connected via HDMI. Same happens if I connect the monitor via USB-C with a converter or natively with a DP cable.

I just realised that there are now two NVIDIA driver branches(?)/versions available since v515. When I created my report I did not noticed this difference. I installed the open kernel version of v520 first. So recently my issue was related with that mostly. Meanwhile I installed the original closed version and with that the inputs working IF I set the GPU handling to on-demand or nvidia. With intel switch this driver also fails and there are no video signals on the outputs. So basically both driver versions affected but somehow the original is partially working at the moment.

Things going wild after v525. The open kernel version does not send any signal. The closed version cause serious continuous lag on external display when the profile is set to on-demand. But when I set it to nvidia it works well. If this is the 21st century and the upcoming year is the year of Linux again then I should forget completely this platform.

I have Lenovo Legion too with GeForce RTX 2060 Mobile and I had the exact same problem. I first updated to nvidia-driver-525-open in the Software & Updates application on Ubuntu 22.04 under Additional Drivers.

What solved this issue of HDMI not being found is choosing the previous driver package NVIDIA driver metapackage from nvidia-driver-515 (proprietary). With this package, everything worked after restarting the machine.

Reverting the nvidia driver to the non-open variant of the same version, i.e. nvidia-driver-530 works for me. I assume there are some issues or lacking features in the open kernel driver variants that break external displays for now.

After installing the proprietary Nvidia drivers (version 455) under Ubuntu 20.10, I am unable to reboot into my system. I simply get a black screen with a blinking cursor, and then absolutely nothing happens.

Another issue I thought of would be that the driver is too old to support an RTX 3070. But checking the build of the drivers on launchpad for 20.10, the release notes literally state that they have 3070 support.

Hello, I have a problem . I reinstalled Garuda because I wanted to start fresh after a mainboard change. So all other parts have run successfully under Garuda before. Unfortunately fixed, I have already had some errors during the reinstallation. Which I have now in painful work, I am unfortunately not an experienced Linux user and only recently changed from windows. Now I wanted to install the proprietary Nvidia drivers. Unfortunately, then the PC no longer boots and hangs when creating the ramdisk. Now I have selected a snapshot before the driver update and I am back on the desktop. But as soon as I reboot it hangs again at the ramdisk. So I assume that I have to restore the old snapshot somewhere? I also found the Garuda Boot repair tool. Unfortunately I have no idea how to use it, couldn't find anything in the wiki. Can someone maybe guide me so that I can properly boot.

It looks like you're still using the nouveau drivers. Did you boot the live boot environment with 'Boot with proprietary drivers' it should install the proprietary drivers during installation instead of needing to install afterwards..

In relation to the snapshot, you must actually tell the system to restore the snapshot after you boot into the snapshot. If you just boot into the snapshot but then reboot, it'll go back to the current filesystem.

But the problem with the Nvidia drives still exists.When I click on hardware configuration via the settings manager and install the proprietary driver from there, the pc hangs on a restart when creating the ramdisc.

i have tried the second variant, then the error occurs. however, i do not receive an error message or anything like that. The installation completes, but as soon as I restart I get stuck when creating the ramdisc.

Unfortunately I couldn't tell you. In the meantime I have tried the procedure again. Now it seems the pc is not even able so load a old snapshot (which worked before). Since 15 minutes the pc is now loading the snapshot .

I can check this, but I do not think that is the reason. I had yesterday after I already had the problems briefly installed windows 11. Here the hardware ran without problems. I formatted the hard drive afterwards of course, so I'm not trying a dual boot currently.

i am unfortunately back to an old snapshot. After deleting the python-pyxdg I tried again to switch to the nvidia drivers. I rebooted and was back to the ramdisc problem. I took 2 bad quality pictures of the messages with my phone to see if I could download the firmware. In the software center I could also find qed-git, aic94xx and wd719x, which were also not installed. The others I have not found yet.

I was only able to boot into Fedora using the nomodeset parameter in grub (I assume because of the missing Nvidia drivers). I ran dnf update and then installed the latest Nvidia drivers from the Nvidia CUDA repository (Later I also tried the RPM Fusion method with the same results).

After rebooting, I tried starting Fedora using the kernel 6.1.14 without the nomodeset parameter and I was able to get to the decryption screen and enter my password. The loading animation starts but the screen goes black before I can see the login screen.

My suggestion is to remove what was installed from the cuda repo and instead enable the rpmfusion-nonfree-nvidia-driver repo and install from there with sudo dnf install akmod-nvidia xorg-x11-drv-nvidia-cuda which will give you both the nvidia drivers and cuda as well.

The nvidia driver from rpmfusion is built using akmods and for that purpose dkms.service is not needed. The fix for the dkms error may be to disable the dkms service with systemctl disable dkms.service then reboot.

Sorry, I'm a bit confused. But am I supposed to select which driver to use in some way? I've only downloaded the ISO and formatted a USB drive that I'm now booting the installer from, so I haven't done anything other than that. But I might be completely misunderstanding your question.

I get the black screen selecting the option "Try or Install Zorin OS (modern NVIDIA drivers)" from the initial menu when booting from the USB. I then instead tried the option "Try or Install Zorin OS (safe graphics)" and that worked. Although when I now try to boot the installed OS from the SSD I installed it to, I get the same problem again with the black screen.

I then instead tried the option "Try or Install Zorin OS (safe graphics)" and that worked. Although when I now try to boot the installed OS from the SSD I installed it to, I get the same problem again with the black screen.

Alright. In my case, it says "NVIDIA Corporation: Unknown" in the top, followed by: "This device is using a manually-installed driver". I remember seeing some message during the installation about detecting devices, could it be that the installer tried installing something based on what devices that were discovered during that?

@Storm I'll try your suggestion, just wanted to provide some info that I found following the official guide for activating Nvidia here on zorin.com. I had a look under Settings > About and under Graphics it says "llvmpipe (LLVM 12.0.0, 256 bits)".

[drm:nv_drm_load [nvidia_drm]] ERROR [nvidia-drm] [GPU ID 0x00000100] Failed to allocate NvKmsKapiDevice

[drm:nv_drm_probe_devices [nvidia_drm]] ERROR [nvidia-drm] [GPU ID 0x00000100] Failed to register device

3a8082e126