Hi all
The PSSM has a GLSL shader inside the implementation. And this makes the
lighting stuff different from default one. Even coloring, .. , transparancy
isn't a easy  task. So to solve such unwished behaviour you can replace the
pssm glsl shader our self. i would copy the org pssm shader from the current
pssm implementation and rewrite those part you need to fix. then you can
pass the new glsl shader thourgh the pssm's shader exchange interface.


        /**
         * you can overwrite the fragment shader if you like to modify it
yourself, own fragment shader can be used
         */
        class OSGSHADOW_EXPORT FragmentShaderGenerator : public
osg::Referenced {
            public:
                /**
                 * generate the GLSL fragement shader
                 */
                virtual std::string generateGLSL_FragmentShader_BaseTex(bool
debug, unsigned int splitCount,double textureRes, bool filtered, unsigned
int nbrSplits,unsigned int textureOffset);
        };

        /** set fragment shader generator */
        inline void setFragmentShaderGenerator(FragmentShaderGenerator* fsw)
{ _FragmentShaderGenerator = fsw;}



i hope this solve your problems.

important to know : (SingleThreaded is a must)
osgshadow --pssm --SingleThreaded




/adrian

2009/4/8 Wojciech Lewandowski <lewandow...@ai.com.pl>

> Bob,
>
> Its hard for me to imagine how one could get the effect of lightening
> instead of dimming. But maybe color material set in vertices
> (GL_COLOR_MATERIAL)  may be involved somehow. This attribute is not handled
> by vertex shaders. But its handled by fixed pipeline.You may look at the
> shaders code ( stored inside osgShadow\src\StandardShadowMap.cpp ). All
> classed derived from StandardShadowMap (Lispsm included) may work bit
> differenet than other techniques because they define Vertex Shader. If your
> fixed pipeline lighting setup is different than what is assumed in SSM
> vertex shader we may see some issues not observable with other techniques
> (which only use Fragment Shaders). SSM vertex shader assumes that we have
> one light and separately compute and pass diffuse and ambient components to
> fragment shader. If ambient is passed separately, fragment shader may
> properly remove only diffuse component  in shadowed areas keeping ambient
> untouched.
>
> Cheers,
> Wojtek
>
> Cheers,
> Wojtek
> ----- Original Message ----- From: "Bob Holcomb" <bob_holc...@hotmail.com>
> To: <osg-users@lists.openscenegraph.org>
> Sent: Wednesday, April 08, 2009 2:34 AM
> Subject: [osg-users] osgShadow and model alpha textures
>
>
>
>  Hello,
>>
>> I'm adding osgShadow to an existing osg project and I'm getting some weird
>> results with PSSM (Parallel Split Shadow Maps).  They don't seem to be
>> respecting the alpha channel of my model's texture.  For example, I have
>> helicopter model who's rotor system is represented by a quad and a texture
>> of the blades which is mostly transparent. When rendering before adding
>> osgShadow it looked fine.  After adding osgShadow the translucent part is
>> white (but the shadows projected on the ground look pretty good!).  This
>> only affects the items under the shadowedScene node, any geometry not under
>> that node is rendered correctly.
>>
>> I've also tried the Light Space Perspective Map technique and found that
>> the self shadowing looks great, but the projected texture on the ground
>> actually makes the ground lighter instead of darker.  If anybody has any
>> info on either of these issues I would also appreciate the help.  Thanks!
>>
>> Cheers,
>>
>> Bob
>>
>> ------------------
>> Read this topic online here:
>> http://forum.openscenegraph.org/viewtopic.php?p=9871#9871
>>
>>
>>
>>
>>
>
>
> --------------------------------------------------------------------------------
>
>
>
>  _______________________________________________
>> osg-users mailing list
>> osg-users@lists.openscenegraph.org
>> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
>>
>>
> _______________________________________________
> osg-users mailing list
> osg-users@lists.openscenegraph.org
> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
>



-- 
********************************************
Adrian Egli
_______________________________________________
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Reply via email to