[osg-users] LiSPSM contibution ?

[Adrian Egli OpenSceneGraph (3D)]

Hi

As you know i tried to implement Parallel Split Shadow Map, may you shadow
experts can review the PSSM as well as i can try to check  LiSPSM as well.

May LiSPSM and PSSM are quite close in concept, both like to optimise the
viewing frustum, would it be possible to integrate the LiSPSM in PSSM or
vise versa?

regards adrian

2008/5/26 Wojciech Lewandowski <lewandowski at ai.com.pl>:

> Hi J-S,
>
> >You say SSM is equivalent to osgShadow::ShadowMap... Is it
> > view-dependent though (as inheritance from ViewDependentShadowTechnique
> > would seem to show)?
>
> Yes it is view dependent. Have you seen some issues with pure
> osgShadow::ShadowMap in multi view configurations ?
>
> > Objects outside frustum can cast shadows.
> > Surely, but did you check those two cases I mentioned? It may just be a
> > bug in certain circumstances... I definitely saw some shadows disappear
> > or be clipped, and it was very obvious. In such small scenes (-3 for
> > example) where no objects are behind the camera and the scene bounds are
> > very small, it shouldn't happen I think...
>
> minLightMargin is exactly for this purpose. I should have set it to some
> positive value. I use following rule of thumb: it should be max length of
> largest casted shadow in the scene. So if objects have 10 units of
> elevation
> I set this distance to 10 x 10 = 100.  For such small scenes as -3 probably
> 10 will suffice.
> You may try to add setMinLightMargin( 10 ); in osgshadow paragraph
> where --lispsm technique is created. There is a catch though. You may be
> hit
> by some assertion code I added today. If you want to try this please update
> ConvexHull.h header (attached here).
>
> I do observe some issues sometimes with point lights / and LiSPSM based on
> MinimalDrawBoundsShadowMap.
>
> > > e) SSM, MSM, MCBSM, MDBSM, LISPSM produce shadows. As I said before I
> wrote
> > > my LiSPSM mainly for large terrain and directional light and believe
> me,
> > > difference between this and SSM is huge in this type of application ;-)
>
> > I'm sure! I'll have to test it with our simulators!
>
> I recomend setting maxFarPlane as well.
>
> >> But if you compare
> >> this with osgShadow::ShadowMap then I would suggest that the problem
> might
> >> be related to using AmbientBias in osgShadow::ShadowMap. Due to
> AmbientBias
> >> uniform and how its used ambient lighting is not correctly computed in
> that
> >> technique.
>
> >That's probably what it is.
>
> I noticed that ambient material property is set to 0,0,0,1 for scene -3. I
> also noticed bizzare issue on GeForce 6200 at my home. Vertex shaders did
> not link. Fixed pipeline indeed produced black shadows. If I swapped shader
> order it started to link and produce smoother albeit still very dark
> shadows.
>
> Cheers,
> Wojtek
>
> _______________________________________________
> osg-users mailing list
> osg-users at lists.openscenegraph.org
> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
>
>


-- 
********************************************
Adrian Egli
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openscenegraph.org/pipermail/osg-users-openscenegraph.org/attachments/20080527/b5d5b27a/attachment.htm>