Yes, I think the shift isn't too great.  The camera being a 'fisheye' with
'angle 160' doesn't help.  Although the shift in position is about the typical
ratio of 1 unit sideways to 30 units object distance mentioned before I also
didn't want it to be too much.  So it should look about like a very large open
area, which it is I guess, not a smaller room-sized area.
Admittedly I saw very little depth in it which is why I shifted the camera even
further than a first trial at it.  Think you could be right and it lacks enough
3D depth nonetheless, the I've always used a 'perspective' camera with something
like 'angle 22.5' (odd number, I know) for them.

Bob

"Sander" <san### [at] stolscom> wrote in message news:38823d3a@news.povray.org...
> I tried to see depth in these Escher pictures, but I can't. Not considering
> the changes (radiosity) to the middle one: are they something you rendered
> with POV? There doesn't seem to be visible the typical POV plasticity about
> them. When I look at two different pictures in detail using PS5, I don't see
> very much difference in position between the objects that should be in front
> and others that should be in the back, as you can observe in clearly 3D
> picture sets. I try to understand what is going on here, but I don't...
> Help!
>
> --
> Regards,
> Sander
>
>
> omniVERSE <inv### [at] aolcom> schreef in berichtnieuws
> 388086b1@news.povray.org...
> > By partly radiositied I mean only one side of the stereopair used it, the
> other
> > is lit with a bright light and has more than default ambient as well.
> Both were
> > darkened somewhat by raising the 'assumed_gamma'.
> > Oh, and this is a combination crossed and parallel image too so all should
> be
> > happy.
> > The Relativity model (Escher stairs I like to call it) was done by someone
> else
> > in Dxf format and I converted it over to Inc with 3DWinOGL (by Thomas
> Baier).  I
> > almost never got the thing positioned right, bet you can guess why.  To
> explain
> > what I was thinking of by doing the non-radiosity + radiosity renders it
> occured
> > to me how well suited stereopairs could be for merging images to cause
> varied
> > effects.  The light and dark changes from one to the other make for an
> > interesting perception.  I've always kind of liked that.  Perhaps a good
> way to
> > describe the possibilities is to think of a window thats both lit and
> unlit or a
> > object which exists in only one image yet shows up in the resulting 3D
> picture
> > seemingly detached somehow.
> >
> > Bob
>
>
>

Post a reply to this message