nvidia GEForce 7300GT under Suse 10.2 slow desktop
Linux - NewbieThis Linux forum is for members that are new to Linux.
Just starting out and have a question?
If it is not in the man pages or the how-to's this is the place!
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.
nvidia GEForce 7300GT under Suse 10.2 slow desktop
I assume as a newb I should post here vs. hardware forum. I installed an nvidia AGP Geforce 7300GT on my Pentium 3.2Ghz Suse 10.2 box using yast online update from the download.nvidia.com site, but I am not sure it went well, even though going back into Yast Software management shows the nvidia-gfx-G01-kmp-default and x11-video-nvidiaG01 checkmarked as installed. Also Yast shows VESA Framebuffer Graphics as Display 1 Card.
"My Computer" window says the display is NVidia but driver is Unknown. Also moving windows around the screen the movement is jerky, they repaint, and scrolling text files is jerky also. The Open GL screen savers don't work, nor does playing video. The resolution and colors look great, but because of the above problems I suspect I don't really have the power of a 512MB video card working for me.
I imagine readers would want me to include all kinds of log excerpts but I'm not sure which would be best to include. Please suggest commands to run and I will do so and cut and paste output into this thread.
I did see some things that didn't look good to me in Xorg.0.log, but I suspect this won't be enough to solve the problem:
(II) v4l driver for Video4Linux
(II) FBDEV: driver for framebuffer: fbdev
(II) Primary Device is: PCI 01:00:0
(II) Loading sub module "fbdevhw"
(II) LoadModule: "fbdevhw"
(II) Loading /usr/lib/xorg/modules/linux//libfbdevhw.so
(II) Module fbdevhw: vendor="X.Org Foundation"
compiled for 18.104.22.1682, module version = 0.0.2
ABI class: X.Org Video Driver, version 1.0
(EE) Failed to initialize GLX extension (Compatible NVIDIA X driver not found)
(EE) FBDEV(0): FBIOBLANK: Invalid argument
Thanks in advance for any help, pointers or diagnostic steps!
suse 10.2 ran extremly slow on my computer and im running 2 8800 ultras in sli with 2 gigs of ram, i tried every tweak there was till i finally dumped it for ubuntu , ubuntu runs lightning fast and it automaticaly installed all my drivers for me with no cd, go ubuntu woooot!!!!!!!!
Hi myzery, I'll keep the idea of switching distro in mind.
Cybertaz, I tried sax2 after su in command window and it just brought up the same screen as yast did, but it did put up a warning box saying "the configuration is framebuffer based" and my system does not support changes, if that means anything. I tried sax2 a variety of ways (based on Googling)...I tried it from as su in command window, I tried it after init 3...I tried it using sax2 -a for automatic configuration...I also tried sax2 -m 0=nvidia...that failed saying I should see /var/log/SaX.log which I did, and it contained the following near the end:
(II) Setting vga for screen 0.
(**) NVIDIA(0): Depth 24, (--) framebuffer bpp 32
(==) NVIDIA(0): RGB weight 888
(==) NVIDIA(0): Default visual is TrueColor
(==) NVIDIA(0): Using gamma correction (1.0, 1.0, 1.0)
(==) NVIDIA(0): Using HW cursor
(**) NVIDIA(0): Enabling RENDER acceleration
(==) NVIDIA(0): Video key set to default value of 0x101fe
FATAL: Error inserting nvidia (/lib/modules/22.214.171.124-0.3-default/updates/nvidia.ko): No such device
(EE) NVIDIA(0): Failed to load the NVIDIA kernel module!
(EE) NVIDIA(0): *** Aborting ***
(II) UnloadModule: "nvidia"
(II) UnloadModule: "ramdac"
(II) UnloadModule: "wfb"
(II) UnloadModule: "fb"
(EE) Screen(s) found, but none have a usable configuration.
Anything else I can try? Thanks for taking the time to reply!
I have done a bit of additional Google research and it looks to me like I, as a newb, have stumbled into a Suse/nvidia/kernel issue people who sound a lot more knowledgeable than I have been struggling with.
I also took your advice and used yast software management to install anything related to Frame Buffers (a number of modules all with fb in their names it turns out, plus their dependencies)
If that assumption is correct, maybe I should change distros (I have considered that just for the sake of my Linux education...I could dual boot another distro, right, and see what I thought of the other flavor of Linux AND see if my graphic card worked better too?)
Or, and I must decide this in the next few days, just exchange the nvidia card for an ATI and see if that works better.
By the way, I have run all software updates to make sure I am not behind on some fix. Also, reading many other Google hits I found that I am supposed to be seeing an nvidia splash screen and I am definitely not, so I think it's in some kind of compatability mode.
Last edited by legacyprog; 09-16-2007 at 12:05 PM.
Further actions I took. I just downloaded/burned and tried an Ubuntu 7.04 Live Desktop CD. It booted OK after some glitches with the video, desktop resolution wasn't very good (1024x768 at 61hz) and I had no ability to change it. But there was no jerkiness to window movement or scrolling. However I didn't seem to have any kind of ability to poke around and see what driver it was using, couldn't get sax2 to run...probably because I'm even more of an Ubuntu or a Live CD newb than a Linux newb in general.
I figured I would try adding an Ubuntu partition to my machine along side Suse to see if that gave me more control/research ability, but when the install app took me to the partition app it said I had a 117 GB reiserfs /dev/sda2 device with unknown space used. I didn't see any way to preserve my suse system and data, perhaps I need to do that from within suse, carving out space for Ubuntu before booting from the Live CD and installing? I can google but if anyone has answer at their finger tips I'd appreciate it.
I have family commitments coming up in a bit so if you don't get a quick response it's because I'll be away for a few hours.
Well, at this point I'm kind of stuck (as a newb, others might not be). I'm not good enough to figure out how to partition the machine to have Suse 10.2 with another Linux distro (there's only one 120gig hard drive on this machine) in order to see if I can get this nvidia geforce 7300 working under another distro, and I've definitely met my match trying to get it to work under Suse 10.2.
So I'll check back once in a while but if I don't hear any further suggestions I think I may just exchange this nvidia card for an ATI one and see if I have better luck. Actually this machine used to have an ancient ATI AIW 7500 in it and this was my idea of an upgrade to a more modern video card, but it's been a nightmare instead (at least for a newb like me).
Thanks for all who tried to (or perhaps may still) help.
Hmmm. Additional googling suggests it has to do with the version of gcc used by nvidia vs the version used by my kernel. If someone knows how I can find out those two values and more importantly how to make nvidia's run file use the right gcc, please let me know. By the way I know from gcc -dumpversion that 4.1.2 is what's installed on my Suse 10.2 box, but I gather that could be different than what was used for the kernel (or not?). Also I saw someone say to look at /proc/version but that's a 0 bytes file. You're probably shaking your head, but hey, I admit to being a newb.
Thanks cybertaz, but I've been down that road. Long ago (or it just seems that way...actually last Saturday, date of my OP above) I went to that exact link that you suggest (http://www.nvidia.com/object/unix.html) and then on that page I clicked on the link for Latest Version IA32 which took me to http://www.nvidia.com/object/linux_d...100.14.19.html and when there I saw the warning about how Suse users should see http://www.suse.de/~sndirsch/nvidia-...ler-HOWTO.html which I did, and then I took their advice to NOT go down the path of recompiling the kernel but instead to just add the HTTP repository download.nvidia.com to my yast installation source list and then to pick the following two packages:
UNLESS I had certain specific cards, and the 7300 which I bought on Saturday wasn't in that list. I believe my kernel flavor is default.
The only thing I noticed as I went back over and copied all that into this reply was that I may not have been at runlevel 3 when I did the yast thing (as suggested at the suse cautionary page above), but isn't that contradictory? Can you be in the yast GUI in runlevel 3? I am willing to give it a shot after I eat my dinner.
I realize I'm taking up a lot of folks' time trying to help me, but I hope that (once the problem is finally solved) people googling may find this thread and it may save them a lot of trouble. Even if it's something stupid I did, somebody else might make the same mistake, huh?
Thanks and/or sorry!
Last edited by legacyprog; 09-19-2007 at 07:42 PM.
Hi cybertaz. Well, as promised I gave it a shot but no joy. Specifically, I did a Ctrl-Alt-F2 and with su did init 3, I then signed back in with normal user and did startx, then went into yast and uninstalled the two nvidia drivers I downloaded before with yast from nvidia's http repository...then I went back into yast and downloaded them again, then I did Ctrl-Alt-F2 to make sure I was still in init 3 which I was...so then I did su and then init 6 to reboot...after rebooting the window movement/scrolling was as bad as ever, so I looked at xorg.conf and xorg.0.log which both showed minutes old timestamps, but otherwise looked the same as I posted above. I still see Framebuffer Graphics with vendor VESA in xorg.conf, though I do notice several mentions of nvidia in xorg.0.log, so I guess that's your point, that the hardware is recognized but the driver isn't there, but it's not clear to me what I need to do beyond what I've done to get the driver loaded.
At various times, based on google hits, I've tried sax2 -m 0=nvidia but that (as I pointed out in the thread above) seems to say that nvidia.ko is invalid, which other googling seems to say has got something to do with versions of gcc. I'm still lost and adrift.
Last edited by legacyprog; 09-19-2007 at 09:49 PM.
Reason: fixed typos
When you download from nvidia you do not need to re-compile the kernel. The drivers will need the kernel source to compile it's module. Installing the driver this way only takes about 5 mins. That is compiling module, installing files and then re-starting X. I find that way is much easier than dealing with the rpms.
It sounds scary to have to use the kernel source, but it is quite easy and the installer does everything. You will need the source for your kernel (search in yast for kernel-source and then choose your flavor ie -smp, -default whatever).
You will need to go to a console session (no xorg) in order to install the driver. I would log-out and then at the login screen goto system->console. This takes you to a command line screen login and go to the file location and run the installer. That is it.
Doing it this way has draw-backs. Everytime you update the kernel, you have to re-run the installer. I have always used this way and have had no problems with it.