On Tue, 8 Aug 2000, J.A. Bezemer wrote:
>
> On Mon, 7 Aug 2000, Dale Scheetz wrote:
>
> > I took this private just to pick your brains ;-)
>
> Cc:'d to debian-x because I've had this problem too + workaround
>
> > Your suggestion on the mouse was right on target. However now I have a new
> > problem. I've had it before and still don't know how to resolve it.
> >
> > When I configure X with XF86Setup the final test of the server works just
> > fine, but when I try to run with the new configuration file, all of the
> > Mode settings are rejected with messages like:
> >
> > (--) SVGA: There is no mode definition named "640X400"
> >
> > For each of the mode lines that actually do appear in the config file.
> >
> > Can you give me any pointers on how to approach this problem?
>
> Enlarge the HorizSync and VertRefresh ranges in /etx/X11/XF86Config by 0.5 to
> both sides (or even larger if that still doesn't work).
>
> XF86Setup seems to calculate possible ModeLines from the entered/selected
> H/V-sync ranges, but it doesn't look at the chipset's clocks at that point.
> Many older videocards have a range of fixed clocks that differ just slightly
> from the "ideal" clock that XF86Setup has calculated. But the slightly
> different clock results in slightly different H/V-sync rates, often just
> outside the specified range. The X server checks this only at "normal"
> startups, i.e. not when XF86Setup is trying it out.
That explians the failure. Choosing a different monitor type in XF86Setup
seems to have fixed the problem ... sort of. (I get a smaller image area
than the full screen and vidtune doesn't help much in resizing it.)
>
> This is especially nasty when you select a "Generic VGA compatible" card
> (VGA16 server) with a "640x480 VGA monitor" and simply aren't able to
> reproduce the very mode that XF86Setup uses :-(
>
There _was_ a clocks line. The server comlained about it, saying something
about it not making sense for this server. Commenting out the clocks line
only changed the values that the server declared for clocks and still
didn't resolve the problem the way choosing another monitor type did.
I guess I'm still a bit confused about how the config file controles this
data...
Anyway, now I can bring up the server and get a screen. Boy is this
machine dog slow! Enlightenment segfaulted on the first attempt, but then
came up OK. This is a really good torture test machine if you can stay
awake long enought to see the next move ;-)
I really appreciate your sharing what you know. This is one of the major
benefits to Libre Software. Mind share is at least as important as code
share. That is what made Debian so attractive to me in the first place,
aside from the fact that it produces a killer distro.
Thanks again,
Dwarf
--
_-_-_-_-_- Author of "The Debian Linux User's Guide" _-_-_-_-_-_-
aka Dale Scheetz Phone: 1 (850) 656-9769
Flexible Software 11000 McCrackin Road
e-mail: [EMAIL PROTECTED] Tallahassee, FL 32308
_-_-_-_-_-_- See www.linuxpress.com for more details _-_-_-_-_-_-_-
