[SOLVED] Proprietary NVidia driver: binary installer vs. SlackBuilds.org package creation
SlackwareThis Forum is for the discussion of Slackware Linux.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
Proprietary NVidia driver: binary installer vs. SlackBuilds.org package creation
Hi,
When I have to configure an NVidia video card on a client's machine (as well as on my own dual-monitor workstation), I always use the proprietary driver provided by NVidia. For the past two or three years, it looks as I've been undecided on how to install it. I'm using the binary installer downloaded by NVidia on about one half of these machines. On the other half, I'm using SlackBuilds.org's build scripts, usually with sbopkg.
I don't see any reason to favour one method over another. What's your take on this? What are the potential pros and cons?
I don't see any reason to favour one method over another. What's your take on this? What are the potential pros and cons?
I personally prefer the slackbuild route as that way everything is done through the same package-manager (and removing nvidia without can be a real pain). The only time I have to deviate from using the slackbuild is for computers that require the legacy-driver. For some reason the slackbuild for the legacy driver has removed the multilib support.
The problem with slackbuilds is that they don't seem to detect which are the good versions for the drivers : they depend on the kernel and of the GPU. For example for the last slackware updates, I had to switch from 340.24 to 340.76 (for the kernel), and cannot go over 340 (for the GPU).
A very good thing would be a slackbuild which detects everything and download the appropriate files.
The problem with slackbuilds is that they don't seem to detect which are the good versions for the drivers : they depend on the kernel and of the GPU. For example for the last slackware updates, I had to switch from 340.24 to 340.76 (for the kernel), and cannot go over 340 (for the GPU).
A very good thing would be a slackbuild which detects everything and download the appropriate files.
Something like Debian does with its "nvidia-detect" script which uses the Nvidia "supported devices" lists to determine the available driver(s) for your GPU and has knowledge of the various distro releases, could be done for Slackware as well: https://anonscm.debian.org/viewvc/pk...82&view=markup
If this would be part of the SlackBuild I would always be using that. Now, I just download the binary from nvidia.com and install it manually.
I always install manually. I've never had a problem with uninstalling either using the blob. But, I have an older card, so I may need a new card, or better yet a new computer. This one will be 8 years old next month.
I perfer the Slackbuild install because it automatically backs up the xorg-mesa drivers. One thing I wish it included was the nouveau-blacklist.conf file rather than the /extra package because the package in /extra causes a version control problem with xf86-video-nouveau with slackpkg.
I think the legacy packages are fine as-is. I too have a GeForce 9800GT O.C. that's showing it's age, but at least my legacy package is fine as-is. However the SBo packages could have a list of supported hardware in the Readme.
That case is very familiar. I wouldn't mind that box, some extra RAM and spare disk drive for nothing looking at the custom ventilation holes.
OT - I have been using the nVidia installer for quite a while. The nouveau diver is always improving, and if you want to play, then the SlackBuild is good. However, nVidia have the intellectual property to get the best out of the hardware, so I run the proprietary driver.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.