I'm installing Linux as a no cost, low resource way to set up a couple of file servers and as a way to teach Linux to some younger kids. The machines are all Pentium 2 266's with 128mb of EDO-RAM. The graphics cards are old ISA cards with VGA out puts, they look to be Powercolor brand, so its possibly ATI. On one of the machines, no matter what distro I try to install, after the setup finishes and it starts loading modules, the monitor goes blank, as if its lost signal. No response, not a thing. I can't get the log in screen to come up. I've tried every distro I have, from openSUSE, ubuntu, kununtu, and gentoo to Fedora. Absolutely nothing. Same response. Oddly enough, doing a bare installation, such as the server install works perfectly. Any ideas on what the problem may be?
Thanks for the help in advance
I'm not too sure how useful this will be or if you have already tried it, but you could always look at trying to boot the system up off a LiveCD. If you've no LiveCD version and you don't want to wait for an hour for one like Ubuntu/Kubuntu to download, you can always grab a copy of FeatherLinux which is 75mb in size. It's a rebuild of Knoppix and is good for this kind of work
I'm sure the experts can chime in and give you the real details 8)
My guess is that X doesn't like the graphics card or your X configuration doesn't like it. That's why the terminal will look okay but the GUI will fail. Post the output of "lspci" and your Xorg.conf and we'll have a look-see.
Whoops! I guess installing X.org might help if you want to use a desktop environment
Also, If you want a leightweight desktop environment that should work better with your older gear, I would advise that you stick with XFCE over GNOME or KDE as XFCE was designed to consume as few resources as possible while still remaining useful.
If you don't want to manually install it, I suggest you check out Xubuntu, which is a version of Ubuntu that comes with XFCE as the defualt desktop environment.
Is this your first time installing a linux OS? This is purely out of curiosity because I haven't seen you visiting this particular forum before.
Your post count and comments indicate that you have been around here long enough to know what you are doing computer-wise, so don't kick yourself because even the experts in any field have brain farts.
I was somewhat aware of his respected status among the CPU forum members, and based on some of the stuff that I've read that he has posted, he is respected with good reason. I was, however, curious if he was new to the whole linux scene as I have never seen him on this board before. That coupled with his statement on either this or a different thread " This is my first linux forum trophy" leads me to believe that he is only just now venturing into the linux world/forum.
Also, I realize that in my last few posts, I am coming off almost as though I were a linux veteran. Although nobody has called me on it, I want to take the opportunity to clarify. I have only been using linux sporadically over the past year, and this year, I would say that my OS usage lies in roughly 75 to 80% linux. I may be new to the scene myself, but I am doing the best I can to learn as much as possible, and I enjoy helping others with issues when I can because I know what it is like to have difficulties.
You've basically solved the issue you were having on your own. That's not my definition of a "n00b". Plus, you learned a few good Linux trouble-shooting techniques in the process, decreasing the chance that you'll head back to this corner of the forums with a problem and more likely to come as an adviser.
Don't be overly confident but never sell yourself short either.
That is really nice of you to offer to do linux benchmarking as well. As for proper benchmarks, people use glxgears to benchmark their graphics even though i heard that it really isn't supposed to be used as a bench. In terms of game benchmarks, you could also try loading up a copy of quake4 or doom 3 and then using its in game benchmark. I had also talked to Linux_0 in the past about linux benchmarking and he mentioned talking about timing how long it would take to compile major things that take a while to compile, like maybe a kernel or openoffice.org or compiling firefox. I am sure that Linux_0 can provide extra suggestions.
When I did a project in my Architecture course on the effects of gcc compiler optimization flags, I used benchmarks that chewed on the processor to try to isolate the effects. As such, I used nbench (a bit old but still pertinent), the POVRay built-in benchmark, and transcode a test mp3 with lame. The latest POVRay is multithreaded and there is a SMP-enabled lame fork that you can use if the system is multi-core (as so many are these days).