hi,
iam using DELL system.it has Graphics card Nvidia GeFORCE FX5200 CARD

I started the Redhat linux 9.0 text installation (typing 'linux text' at the prompt). Otherwise, my sreen died.
2) I skipped the X configuration part of the setup
3) I downloaded nvdia driver NVIDIA-Linux-x86-1.0-5336-pkg1.run from nvidia site installed the driver with the sh NVIDIA..... command
4) I copied and (renamed as XF86Config) the nVidia supplied XF86Config-sample file to /etc/X11/

then i started the X server by startx command then it is giving error as foolw as
X10 fatal 10 error 104 ( connection reset by peer) an Xserver " :0,0" after 0 request (0 known Processed ) with 0 events remaining.

this above the error iam getting when iam starting Xserver
any one please help me


regards,

pinkponky: I'd recommend trying the latest version of the drivers from http://www.nvidia.com/

That said there seem to be quite a few people having problems with the latest, I managed to install it, but it just froze on starting X (search the forums linked from the driver download page).

One key problem that I encountered was that I use a custom kernel, and in order for the driver to build properly it needs the sources. Not a problem at all, but I discovered that you should NOT type 'make clean' at any stage of the kernel compilation/installation as it removes a number of required files (when trying to compile the nvidia driver I kept on getting messages saying that kernel sources could not be found despite the fact that they were installed (in /usr/src/linux under slackware).

I managed to get version 6111 working no probs, however I'd appreciate tips on configuring X.org to work. If anyone knows of any significant variations/differences I'd be interested to know details.

Thanks

slack---line

I am here to post a problem and a solution for anybody out there who might encounter the same thing.
After installing and uninstalling the newest nvidia driver (suse 8.2, kernel 2.4.20) several times, it didn't seem to work. The nvidia logo would come up then fail about 4 times before dropping me into text mode whenever I tried to run startx, and startx would report that module glx and libGL.so could not be loaded.
What I discovered was that several important symlinks had not been created correctly by the installer. The support document linked below gave the solution to this. I had to manually check on several different *.so and *.o files, and change the symlinks through the ln command for 2 of them. Hopefully this info will help somebody... it was a really tough problem.

http://nvidia.custhelp.com/cgi-bin/n...p?%3C?%20print(%22p_sid=$p_sid&p_lva=$p_lva&p_li=$p_li%22)%20?%3E&p_faqid=141&p_created=1099953549
(copy-paste the whole url into your location bar, the link is kind of screwy)

Anyhow, I'm glad it works now.

hello
if you are not downloading from your linux computer but have to transfere it, what directory/file do you save t he device driver to?

In sarge, the runlevel 3 dowsn't kill the X server at all, It must be runlevel 1...
After this, you will notice that there is no existing precompiled module matching your kernel (no matter 2.4.27 or 2.6.8-2-686-smp, is the same)
then you proceed to install the your kernel headers and do ln -s /lib/modules/(uname -r) /usr/src/kernel-headers-(uname -r) which link the kernel headers
to this location where the nvidia drivers will look for them, and when you try again to install them, the installer tells you that
/lib/modules/2.6.8-2-686-smp/build/include/linux/modversions.h doesn't exist since you didn't configure the kernel headers, you configure them and it still
doesn't exist, so you copy this file whic is created in another directory to the place that the drivers want it to be, execute the installation again and now it says
it couldn't load nvidia.ko since the kernel headers are not the headers of the kernel you are using, so I don't know what else to do...
Hope anyone using debian sarge gets to do it succesfully, I have no more ideas....

newb.
FC3 just updated to latest kernel about 10 mins ago.
GeForce FX5200
Ok, I had trouble editing the Xorg.conf file so I'm just going to do that after I finish w/ the install
So the next step is to do "init 3" in terminal. So I do that. Screen goes black and as far as i can tell it stays that way. I noticed that if I press "s" it prints the following on the screen:
^the ok's are supposed to be lined up(like in the boot up)^

Looks like when you boot to me, but it only does those 4 things. The underline looks like a cursor to me, but it won't let me type anything. I can't even toggle num lock. The only way I can get it to respond is to do Ctrl+Alt+Del and it will reboot. I did notice in one of the above posts that someone mentioned updating the Kernel and having 2 of them now and having this same problem(he/she said just black screen tho). I currently dual-boot b/w WinXP(games) and FC3 but after I updated the kernel there were 2 options for booting into linux(each kernel). Would this cause a problem? Anyone got any ideas? This slow graphics processing is killing me.


A few notes:
-When I went to change Load "dri" to Load "glx" it was already changed and I didn't do it.
-Is it 100% neseccary to do the Driver "nv" to Driver "nvidia" thing? When I do that and get the error I described above and have to reboot it always complains that it doesn't have a driver called nvidia installed and takes forever to reset and boot.
-I don't know why(maybe b/c I've got a 64-bit) but I don't have an XF86Config file. I have to edit a file called Xorg.conf. This took me about a day to figure out, so if you can't find XF86Config then look for Xorg.conf

When I am installing the NVIDIA 32 bit driver for a Quaddro 3700 acclerator card, I get a message that says that CC is not in the path. Check to see if GCC is installed, if not install GCC. If GCC is installed , then put CC in your path. OK I checked and GCC is installed, so where and in what path do I put CC? This is URGENT for a medical device if anyone can help please let know!!!! Thanks.