Hi @marioroy
I have not been active in the forum because of heavy workload recently. Thought I would let you know that the 530 driver is working fine on my machine. The ‘regression’ issue is not troubling mine-I have a 4k LG tv and a 144hz 1080p monitor. And aren’t we lucky to have a rolling release distro that is cutting edge and fast?
Just for your amusement, I asked ChatGPT 4 (with bing search) to compare your script to the official guide on installing the driver 
and response was;
The method provided in the GitHub script and the official guide to installing the Nvidia graphics driver have similarities and differences.
Both methods have pre-installation requirements:
GitHub Method:
Requires a recent Clear Linux image (version 39110 or newer)1.
Depending on the Clear Linux release, the open-source nouveau driver might not work with recent Nvidia graphics cards (3000 series or newer). Hence, the OS should be installed in text mode-1eon the boot screen and prependnomodeset 3with a space to the list of kernel arguments. The OS will boot into multi-user mode and prevent the nouveau driver from loading. Press enter to boot the OS. Instructions are provided on the screen for running the installer.\n\nDuring setup, remember to enter[A]Advanced options. Select \"Kernel Command Line\" and enternomodeset 3to \"Add Extra Arguments\". Optionally, go back to the prior screen and choose thelts(recommended) ornative kernel under \"Kernel Selection\". Also disable automatic OS updates if desired","pub_date":null}}.
During setup, the user should select "Kernel Command Line" and enter nomodeset 3 to "Add Extra Arguments". The user can also select the lts or native kernel under "Kernel Selection"-1eon the boot screen and prependnomodeset 3with a space to the list of kernel arguments. The OS will boot into multi-user mode and prevent the nouveau driver from loading. Press enter to boot the OS. Instructions are provided on the screen for running the installer.\n\nDuring setup, remember to enter[A]Advanced options. Select \"Kernel Command Line\" and enternomodeset 3to \"Add Extra Arguments\". Optionally, go back to the prior screen and choose thelts(recommended) ornative kernel under \"Kernel Selection\". Also disable automatic OS updates if desired","pub_date":null}}.
The user is advised to consider their specific needs, such as running Blender or using a Nvidia Optimus laptop, which would require the Nvidia driver 520, or if they are using a high refresh rate display or high resolution, they should not install 530.41.03, but rather choose 525 due to a regression with the 530 driver1.
Official Method:
Verifies the system has a CUDA-capable GPU.
Verifies the system is running a supported version of Linux.
Verifies the system has build tools such as make, gcc installed.
Verifies the system has correct Linux kernel headers2.
The installation process also differs:
GitHub Method:
The user has to clone the GitHub repository and run the pre-installer script with the update argument if needed, to refresh the configuration files under /etc/-1updateargument is needed on prior Clear Linux installations or to refresh the configuration files under/etc/. Run the pre-installer script subsequently, without an argument, to switch the target from graphical to multi-user (text-mode).\n\n $ bash ./pre-install-driver help\n Usage: pre-install-driver [ update ]\n\n $ git pull\n $ bash ./pre-install-driver update\n $ bash ./pre-install-driver\n\nNext, run the driver installer script. Running the LTS kernel? Choose any driver in the list. Planning on running the native kernel or GeForce RTX 4000 series? Specify 525minimally or latest. Or provide the location of theNVIDIA-Linux-x86_64-*run-file. Do not forget to run thecheck-kernel-dkmsscript. That will check each kernel and involvedkms to auto-install the NVIDIA modules, if needed.\n\nDriver Version \nlatest 525.116.04 \n530 530.41.03 \n525 525.116.04 \n520 520.61.05 \n\n $ bash ./install-driver help\n Usage: install-driver latest|530|525|520|<valid_pathname>\n\n $ bash ./install-driver 525\n $ sudo reboot\n\nRun the check script if you have multiple kernel variants on the system.\n\n $ bash ./check-kernel-dkms","pub_date":null}}.
Then, the user has to run the driver installer script, choosing the driver version suitable for their system, or provide the location of the NVIDIA-Linux-x86_64-* run-file. The check-kernel-dkms script is run afterwards to check each kernel and involve dkms to auto-install the Nvidia modules, if needed-1updateargument is needed on prior Clear Linux installations or to refresh the configuration files under/etc/. Run the pre-installer script subsequently, without an argument, to switch the target from graphical to multi-user (text-mode).\n\n $ bash ./pre-install-driver help\n Usage: pre-install-driver [ update ]\n\n $ git pull\n $ bash ./pre-install-driver update\n $ bash ./pre-install-driver\n\nNext, run the driver installer script. Running the LTS kernel? Choose any driver in the list. Planning on running the native kernel or GeForce RTX 4000 series? Specify 525minimally or latest. Or provide the location of theNVIDIA-Linux-x86_64-*run-file. Do not forget to run thecheck-kernel-dkmsscript. That will check each kernel and involvedkms to auto-install the NVIDIA modules, if needed.\n\nDriver Version \nlatest 525.116.04 \n530 530.41.03 \n525 525.116.04 \n520 520.61.05 \n\n $ bash ./install-driver help\n Usage: install-driver latest|530|525|520|<valid_pathname>\n\n $ bash ./install-driver 525\n $ sudo reboot\n\nRun the check script if you have multiple kernel variants on the system.\n\n $ bash ./check-kernel-dkms","pub_date":null}}.
Optionally, the user can also install the Nvidia CUDA Toolkit. The "auto" argument is preferred, which will install the version suitable for the display driver1.
Official Method:
Nvidia provides .run installer packages, which the user can download and then install by running the .run file with sudo privileges2.
Nvidia also mentions the option of installing the drivers through a Package Manager, which interfaces with your system's package management system. The downloaded package is a repository package, which informs the package manager where to find the actual installation packages, but doesn't install them itself2.
Overall, both methods aim to achieve the same goal but take different approaches. The GitHub method provides a more detailed and customized approach for Clear Linux OS users with certain specific requirements, while the official Nvidia guide provides a more general method applicable to a wider range of Linux distributions.
Wish you a good weekend.
