Current Xorg behavior is unfriendly at best

[Arthur Pemberton]

On 9/18/07, Adam Jackson <ajackson at redhat.com> wrote:
> On Tue, 2007-09-18 at 11:25 -0500, Arthur Pemberton wrote:
> > On 9/18/07, Adam Jackson <ajackson at redhat.com> wrote:
> > > If the monitor is off, I can't talk to it, so I can't ask it what it can
> > > do.  Note that this is something of a lie, most monitors made since, oh,
> > > 2000, will merely look like they're asleep but will still respond over
> > > DDC if they have a power cord plugged in.
> > >
> > > But if I can't talk to it, I'll use what's in the config file.  Do you
> > > have an example of X not doing so?
> >
> > Yes... the most frustrating part of all. As I've been trying to
> > communicate, xorg does _not_ use my xorg.conf when the monitor is off,
> > and instead uses something on the scale of 800x600 which is so large,
> > the necessary widgets are off screen.
>
> Perhaps I wasn't sufficiently clear about this.  If you could feed me an
> X log and config file from this failure condition, I might be able to
> figure out why it looks like we're ignoring your settings.

I will get you the logs once I get back to the machine, my xorg.conf
has already been provided.

> The likely explanation is that your Monitor section is either missing,
> or doesn't contain any information about the sync ranges and etc.  Which
> means it doesn't describe your monitor at all, so we have to assume
> something conservative like 800x600.

Am I required to manually put in my Monitor section? If so, then going
down to 800x600 is the default behaviour.

> If your favorite app doesn't fit
> in that size, then it's violating the Gnome HIG to begin with.

In this case, it is the login screen, KDM

> It's generally not sensible to try to infer what monitor capabilities
> the user meant based on what modes they asked for.  If I ask for
> 2048x1536 on a monitor that can't do it, I might be able to set up the
> card to scan out that timing, but the monitor's just not going to sync,
> and it's _much_ better to fail in a way that the user can see and
> interact with, than fail to black.
>
> - ajax

Well it seems like many of the current assumptions result in cases
where the end result is "unfriendly", considering that my hardware
setup isn't exotic, and my software (in terms of Xorg) is near
vanilla.

-- 
Fedora 7 : sipping some of that moonshine
( www.pembo13.com )