Nvidia GeForce GT 720 - Will it work?

10 réponses [Dernière contribution]
Geshmy
Hors ligne
A rejoint: 04/23/2015

It's already installed and I see it is well supported by nouveau and I was hoping to avoid the rolling back to an earlier kernel. So far it's worse than the Radeon card I had giving me 1028 X 768 resolution for a 1920 X 1080 monitor. It works fine in Fedora. Any suggestions?

leny2010

I am a member!

I am a translator!

Hors ligne
A rejoint: 09/15/2011

It's shown as supported with experimental reclocking on the nouveau website. Move to a more recent kernel, when I ran nVidia GPUs I always used jxself's forward kernel for best nouveau kernel driver support but take your pick from our wiki page

https://trisquel.info/en/wiki/update-linux-libre-kernel

Geshmy
Hors ligne
A rejoint: 04/23/2015

Hey leny2010, that looks great!
I will give that a try in a bit. I hadn't imagined I could run a 4.x kernel in Trisquel. I am using Belenos, hopefully that won't make a difference. I will report.

Geshmy
Hors ligne
A rejoint: 04/23/2015

I did follow instructions right down the line (didn't need the 'little bash function.') Ran $sudo apt-get install linux-libre64 linux-libre64-header and $ sudo update-grub and rebooted. It looks like it is going to work, though I see a message for a moment something about 'nouveau fault a' but none the less it shows that Trisquel splash screen which I think might be showing at the right resolution but then it just goes white.
Any ideas?
I attached text file with configs and Xorg.0.log.old.

Pièce jointeTaille
NouveauFailNotes.txt 28.13 Ko
Magic Banana

I am a member!

I am a translator!

Hors ligne
A rejoint: 07/24/2010

Nothing wrong in the files you attached...

Geshmy
Hors ligne
A rejoint: 04/23/2015

Haven't given up yet. I wonder if I have conflicts between packages and could benefit from reinstalling and going straight to the jxself 4.x kernel as a starting point. I did make a very small partition for /var/log and recieved a message that it was almost full. Mostly was kern.log (76 MB) which was largely filled up with something like this:

May 21 14:13:55 Trisqueltower kernel: [ 46.225317] nouveau E[ PGR][0000:01:00.0] FECS ucode error 2
May 21 14:13:55 Trisqueltower kernel: [ 46.225349] nouveau E[ PGR][0000:01:00.0] FECS ucode error 2
May 21 14:13:55 Trisqueltower kernel: [ 46.225379] nouveau E[ PGR][0000:01:00.0] FECS ucode error 2
May 21 14:13:55 Trisqueltower kernel: [ 46.225410] nouveau E[ PGR][0000:01:00.0] FECS ucode error 2
May 21 14:13:55 Trisqueltower kernel: [ 46.225442] nouveau E[ PGR][0000:01:00.0] FECS MTHD subc 1 class 0x80011698 mthd 0x1698 data 0x00001000
May 21 14:13:55 Trisqueltower kernel: [ 46.225473] nouveau E[ PGR][0000:01:00.0] FECS ucode error 2
May 21 14:13:55 Trisqueltower kernel: [ 46.225505] nouveau E[ PGR][0000:01:00.0] FECS MTHD subc 1 class 0x80011698 mthd 0x1698 data 0x00001000
May 21 14:13:55 Trisqueltower kernel: [ 46.225527] nouveau E[ PGR][0000:01:00.0] 502000 - stat 0x00000000 0x00000000 0x00000002 0x00000000
May 21 14:13:55 Trisqueltower kernel: [ 46.225568] nouveau E[ PGR][0000:01:00.0] FECS MTHD subc 1 class 0x80011698 mthd 0x1698 data 0x00001000
May 21 14:13:55 Trisqueltower kernel: [ 46.225590] nouveau E[ PGR][0000:01:00.0] 502000 - stat 0x00000000 0x00000000 0x00000002 0x00000000
May 21 14:13:55 Trisqueltower kernel: [ 46.225623] nouveau E[ PGR][0000:01:00.0] 502000 - stat 0x00000000 0x00000000 0x00000002 0x00000000
May 21 14:13:55 Trisqueltower kernel: [ 46.225653] nouveau E[ PGR][0000:01:00.0] 502000 - stat 0x00000000 0x00000000 0x00000002 0x00000000
May 21 14:13:55 Trisqueltower kernel: [ 46.225685] nouveau E[ PGR][0000:01:00.0] 502000 - stat 0x00000000 0x00000000 0x00000002 0x00000000
May 21 14:13:55 Trisqueltower kernel: [ 46.225716] nouveau E[ PGR][0000:01:00.0] 502000 - stat 0x00000000 0x00000000 0x00000002 0x00000000
May 21 14:13:55 Trisqueltower kernel: [ 46.225747] nouveau E[ PGR][0000:01:00.0] 502000 - stat 0x00000000 0x00000000 0x00000002 0x00000000
May 21 14:13:55 Trisqueltower kernel: [ 46.225787] nouveau E[ PGR][0000:01:00.0] FECS ucode error 2
May 21 14:13:55 Trisqueltower kernel: [ 46.225818] nouveau E[ PGR][0000:01:00.0] FECS ucode error 2
May 21 14:13:55 Trisqueltower kernel: [ 46.225849] nouveau E[ PGR][0000:01:00.0] FECS ucode error 2
May 21 14:13:55 Trisqueltower kernel: [ 46.225880] nouveau E[ PGR][0000:01:00.0] FECS ucode error 2
May 21 14:13:55 Trisqueltower kernel: [ 46.225913] nouveau E[ PGR][0000:01:00.0] FECS MTHD subc 1 class 0x80011698 mthd 0x1698 data 0x00001000
May 21 14:13:55 Trisqueltower kernel: [ 46.225944] nouveau E[ PGR][0000:01:00.0] FECS ucode error 2
May 21 14:13:55 Trisqueltower kernel: [ 46.225974] nouveau E[ PGR][0000:01:00.0] FECS ucode error 2
May 21 14:13:55 Trisqueltower kernel: [ 46.226006] nouveau E[ PGR][0000:01:00.0] FECS MTHD subc 1 class 0x80011698 mthd 0x1698 data 0x00001000

davidnotcoulthard (non vérifié)
davidnotcoulthard

How is it working now (looking to get an Nvidia card myself....)

Darksoul71
Hors ligne
A rejoint: 01/04/2012

@davidnotcoulthard: AFAIK older NVidia cards such as a 9X00 or 8x00 have always worked better with noveau then newer cards.

The 610GT in the PC of my daughter was unstable as hell with Nouveau on Ubuntu 14.04 and the 720GT I was testing in another PC didn't run stable as well.

YMMV though depending on the kernel and version of Nouveau of course.....

willieaames
Hors ligne
A rejoint: 10/15/2014

This Card is a great value, have it running 3 monitor (L-1290x1080, C-1260x1440 and R-1024x768). Monitor are not "spanned" just extended desktop.
Very quiet, easy install, add monitors one-at-a-time during configuration and reboot for each monitor.

Darksoul71
Hors ligne
A rejoint: 01/04/2012

Interesting....the 720GT I had for testing in a spare PC of me running Ubuntu 14.04 x64 was very unstable with Noveau.
OpenGL always crashed and accelerated video playback showed artifacts. Are you experiencing similar things ?

Geshmy
Hors ligne
A rejoint: 04/23/2015

Sorry I went dark for so long.

Well, the 720 GT won't earn my recommendation! I couldn't get it to work.
During boot up there was always a message about frame buffer hardware conflict between noveau fb and vesa fb that concluded kernel would remove the 'generic driver.' It would hang for a minute or two and then finish booting up to the wrong resolution. One day I uninstalled the Vesa drivers and voila - new message re 'Unknown chipset Keppler chipset' and then boot process stopped - no more Trisquel. I got that far with leny2010's advice re a forward kernel and whatever else I could come up with re noveau.

I'm afraid it sort of broke me and I haven't been able to bring myself to mess with much of anything tech related for a while. My PC has Centos with the same resolution issues and I have been using that for the minimal personal business stuff I do (including logging in to the forum today) and Fedora with nvidia blobs installed so I can watch youtube or whatever comfortably. I use the laptop with debian for email as Enigmail is working on it.

I read that jxself has a de blobbed kernel that may work with the original Radeon card I had in the PC so when I am recuperated enough and have some gumption again, I'll reinstall that card and start over. I do really like Trisquel and especially enjoy the forum.