I've found the fixed-function uniforms to be one of the more fragile
aspects of GLSL as far as driver support goes.  It's too bad since
they are useful for maintaining fixed-function fallback paths for
something like shadow mapping.  However, it seems that the IHVs are
pushing to get rid of them altogether in the OpenGL 3 lean and mean
profile, a la OpenGL ES shading language.  If you look at the ES
shading language spec on the Khronos site, you'll see a strikethough
on all but the most fundamental builtins beginning with gl_.

Leif

On 8/28/06, Botorabi <[EMAIL PROTECTED]> wrote:
hi,

i had totally missed that part of math concerning the multiplication of the 
inverse modelview matrix; thanks for the explanation Leif :-)

a patch has been prepared for VRC fixing the shadow problem and will be 
released next days (see the project site for more details 
http://yag2002.sf.net).

some people reported that on ATI cards the shadows look a little strange. hehe, 
it seems that i will have to patch the patch next time in order to work-around 
an ATI driver bug.

wow, sometimes it can be really frustrating for a small open-source team facing 
with such problems, as we are limited in hardware equipments and cannot test 
the app on various hardware platforms; and the users report bugs once the 
release is out, not during the pre-release testing phase, hehe :-P

cheers
boto



robert wrote:
#############
Hi Mike,

On 8/27/06, Mike Weiblen <[EMAIL PROTECTED]> wrote:
>
> Hi,
>
> Just curious, you're already using GLSL, why bother at all with texgen
> suff?
> texgen is so GL 1.x ;-)
>

As Leif mentions eye linear tex gen mulitplies by the inverse of modelview
matrix before applying, something you don't get for free for uniforms.

Also TexGenNode is currently built around osg::TexGen rather than uniforms.

Longer term what we need is a class of uniform that behaves like glTexGen.
We can implement this on the OSG side.

Robert.
_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/

_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/

Reply via email to