From ArchWiki
Revision as of 14:15, 1 March 2014 by Silverhammermba (talk | contribs) (Why the Different Package Versions?)
Jump to: navigation, search

XRandR support

As far as i understand it the recent drivers support XRANDR, which is probably much better than Xinerama/Twinview. Should we remove the Xinerame/Twinview instructions alltogether and just mention to use the standard XRandR methods for multiscreen setups?

NVoption Online

NVoption Online Version - great tool to make tv-out easy and fast

[I'm using gmplayer with gl and twinview] [1]


Maybe someone can put this in better words: Logging out, or switching to a different terminal using ctrl+alt+F<2-9> consistently resulted in a black screen, and killing Xorg with ctrl-alt-backspace resulted in a terminal screen with only the top line visible. It turned out that a 'vga=773' added to kernel line was the cause of this. After removing that the problem was solved. Probably something to do with KMS? B.t.w. I have only used x with 'startx', so possibly specific for that way of starting X. rwd

Was it with this driver or nouveau? The proprietary drivers don't have KMS. thestinger 17:42, 13 December 2010 (EST)

This was with the proprietary driver. I originally had put vga=.. because it made gave the bootup screen a higher resolution, and because the beginners guide mentions it ( Apart from leaving out the vga option, I discovered that setting it to the native resolution as explained on fixes the black screens as well. Maybe a warning for using vga= option with with proprietary drivers would be useful.rwd

Well the thing is that vga= is meant for the proprietary drivers only - open source drivers already set the native resolution without a vga command. It can be removed from the beginners' guide though, since it breaks open source drivers, and if the card doesn't support the vga command, it breaks the closed source ones too. thestinger 20:13, 13 December 2010 (EST)

custom kernel

The package changed for kernel 3.0 and the instructions no longer work. Please fix this. Z.T. 09:14, 23 November 2011 (EST)

'/dev/nvidia0' Input/Output error... suggested fixes

Can anyone verify that the BIOS related suggestions work and are not coincidentally set (either automatically when changing the IRQ or turning off ACPI) while troubleshooting? I have found little information that confirms any of the suggestions would work. The file permissions thing seems to be completely unfounded and never works (as noted in the article) that I could find. It would probably be a good idea if we cleaned out items that have not been verified to work. For my setup I was having this error and none of the items in the wiki nor the many file permission search results worked. -- click, them so hard 19:16, 4 March 2012 (EST)

I've added an Accuracy template, please next time add it yourself so that discussions like this are more visible. -- Kynikos 05:40, 6 March 2012 (EST)


I think the "Installing" section is a little ambiguous and could use a bit of rewording. Because the steps are numbered, and little indication is given otherwise, it is implied that you need both the packages named like nvidia-173xx, and the regular nvidia packages. I don't actually have my nvidia drivers working properly, so maybe I'm misinterpreting this, but if I'm right in assuming that you need either the specifically named drivers like nvidia-173xx or the plain ol' nvidia drivers, step 2 needs to be reworded. I would suggest displaying two separate [code] blocks, one with # pacman -S nvidia-173xx nvidia-173xx-utils, and the one that's there now. Then make it explicitly clear that you need to do one or the other, not both. --Sotanaht 18:45, 17 May 2012 (EDT)

Oh, I forgot that the nvidia-173xx drivers were not in the official repos. Scratch the part about including the command for installing that. I still think it's important to make clear that people using the nvidia-173xx drivers do not need the regular nvidia drivers. Also make it clear that people using the regular nvidia drivers do not need any nvidia-XXXxx drivers. --Sotanaht 18:49, 17 May 2012 (EDT)

Well, I've never had to use the old Nvidia drivers, but the note says that the old modules don't support Xorg 1.11 (Arch provides 1.12 now). Unless the situation has changed, those drivers are useless unless you also write instructions on how to safely downgrade Xorg. Please correct me if I'm wrong. -- Kynikos (talk) 11:27, 19 May 2012 (UTC)

Bad performance, e.g. slow repaints when switching tabs in Chrome suggestion broke emerald/compiz

Firefox performs quite poorly for me, so I tried this suggestion and it ended up breaking my WM. All new window borders changed to solid white and would not move around. Can someone else confirm? If so there should probably be a note or amendment to the suggestion. Biltong (talk)

Run a test

There is confusing paragraph saying You can run a test to see if the Xorg server will function correctly without a configuration file.. IMHO, it should be clarified what kind of test the author has in mind, an exact command would be helpful. Currently, this suggestion is simply confusing, especially to less experienced users. --Mloskot (talk) 19:52, 26 November 2012 (UTC)


Several of the commands which are suggested to be run with nvidia-xconfig (such as nvidia-xconfig --twinview) don't work with the current nvidia packages in the repository. I just went through setting mine up so I intend to clean up the ones that I can from my experience. Some don't seem to have a 1:1 replacement (there is a --dynamic-twinview argument now; is that the same as --twinview was?). Teh (talk) 13:10, 20 June 2013 (UTC)

Why the Different Package Versions?

Despite having a 5 series card, I ignored the wiki and installed the normal nvidia driver rather than the one from AUR. Everything has been working fine so far and I get fine framerates in games. What is the reason for recommending different drivers for different cards? If installing the regular driver is a viable option for non-8-series cards, it should be mentioned as well. Silverhammermba (talk) 02:37, 24 January 2014 (UTC)

What graphics card do you have exactly? Which package did you install? Which AUR package do you think you should have installed? -- Karol (talk) 09:45, 24 January 2014 (UTC)
These driver are proprietary which means users must count on NVIDIA. The latest code branch will not be tested on old cards that End of life. So the basic function may still work but no one could guarantee the quality. This is the main reason why open source driver is prefered.--Fengchao (talk) 07:18, 5 February 2014 (UTC)

I can also vouch for using the nvidia package rather than nvidia-304xx - I gave a GeForce 650 ti, which is listed as a 304xx card. The 304xx packages, however, do not work with the latest CUDA toolkit. I upgraded to the nvidia package and it works perfectly - 3d acceleration, CUDA, etc. I think that since nvidia changed how they package their Linux drivers, users no longer need to worry about grabbing specific package versions. I'd be all for updating this, as it initially misled me. rdahlgren (talk) 00:54, 19 February 2014 (UTC)

Please tell me which part of the wiki exactly says you should be using nvidia-304xx with this card, because certainly doesn't. -- Karol (talk) 01:18, 19 February 2014 (UTC)

Indeed, I did not originally understand the distinction between the 5 and 500 series. Since I'm clearly not the only person who made this mistake, perhaps there is a clearer way to word the install section? Perhaps moving the tip about the legacy card list up under the lspci command would emphasize that you need to look up your card in one of those lists to determine what driver you need. Silverhammermba (talk) 14:15, 1 March 2014 (UTC)