Hello Wojtek, Adrian,
Lets imagine following case: suppose that such the
same light will get used by lighting and shadow computation. It may
bring some discrepancies if happens to be preceded by some
trransformation. When used with lighting it will be premultiplied by
inverse model view but when used by shadowing it will be not
premultiplied. So the lighting direction and shadow casting direction
may get out of sync. Using this properly would require some knowledge
and discipline from user. But it may be still worth adding. I would be
grateful If Robert or others could join this discussion and say what
they think.
Personally, I think the lighting computations for shadowing should be
the same as for any other OpenGL application. A light source is
positional state, you can't just assume it will always be at the top of
your graph with no transforms above it.
Even if it might be a valid assumption in one case, it's not in general
because as Wojtek says, if you ever have the case where there is a
transform above the light source (which is also perfectly valid from an
OpenGL standpoint) then the scene lighting and shadows won't match.
I think PSSM should be fixed to handle this correctly. At a minimum, any
such assumptions should be well documented instead of having to read the
code to find out...
J-S
--
______________________________________________________
Jean-Sebastien Guay [EMAIL PROTECTED]
http://www.cm-labs.com/
http://whitestar02.webhop.org/
_______________________________________________
osg-users mailing list
[email protected]
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
