Re: [Jmol-users] desktop graphics on linux [Was: povray on linux]

[Tim D. <stu...@ma...>]

Am Dienstag, 6. Juli 2004 11:33 schrieb Miguel:
> > Am Montag, 5. Juli 2004 18:31 schrieb Miguel:
> >> Tim wrote:
> >> > By the way is there a command-line option for adjusting the
> >> > window size at start-up?
> >>
> >> No, but that is a very good idea.
> >>
> >> Q: What do you think about a user-preference instead of a
> >> command-line option?
> >
> > As long as it's scriptable (e.g. for batch jobs) it's fine for
> > me. Window-resizing is the only way for zooming that effects
> > povray, that's why I'd like to automate it.
>
> Is it the case that the only thing you are trying to automate is
> povray size?

Yes, that's right.

> Now that I understand what you are trying to accomplish I see that
> there are a few problems.
>
> povray output can be completely independent of the jmol window
> size.
>
> We could also change the povray code so that you could pass
> arbitrary parameters to it. That way, you could choose the size of
> the povray window independent of the Jmol window size.
>
> The advantage would be that it would be more flexible.
>
> The disadvantage would be that the povray output would not be the
> same size as the Jmol window.

Is this really a disadvantage? The default setting could be the Jmol 
window size, so there wouldn't be a difference as long as the user 
doesn't change it. A "fixed ratio"-option for width:hight would be 
useful of course. 
All I'm really looking for is a zoom function which effects the povray 
output to get a higher resolution there. The way to accomplish this 
is secondary to me. But I think zoom for povray would be a great 
improvement.

> Another problem ... we can set the Jmol application window size in
> pixels. But that is not the same as the viewer window size ... the
> actual display size. The Jmol application window size will include
> the menus, buttons, and window decorations. The difference between
> the app window size and the viewer size is platform and
> user-setting dependent.

Okay, but nevertheless it makes Jmol more customizable. Usually the 
difference between different platforms should a few pixels but there 
might be cases you want to have a look at a lot of structures at the 
same time, then you need a lot of small windows, but in other cases 
you might want to compare just two structures or examine just one, 
then the standard size or near-fullscreen would be nice. I don't 
think it needs to be exactly the same pixels on all platforms but the 
same magnitude of pixels