[vtkusers] VolumeRayCastMapper changes

[Lisa S. Avila]

Hello,

I am in the process of changing the vtkVolumeRayCastMapper, and I have a 
question for the group - is anyone out there using the RayBounder? I know 
Bill Lorensen uses it in several examples where volumes and surfaces are 
intermixed. Is anyone else using it? If not, I would like to consider 
removing this functionality from the vtkVolumeRayCastMapper.

Here are the advantages to removing it:

1) This will make the different rendering strategies more interchangeable 
(ray bounding is not supported for the vtkVolumeTextureMapper2D or the 
vtkVolumeProMapper)

2) This will simplify the ray casting code

3) This will eliminate the problems that occur because the ray bounder 
depends on OpenGL storing a Z value in the depth buffer and being able to 
correctly read the depth buffer. Many PC video cards / drivers have 
problems with at least one of those requirements and hence ray bounding 
fails on those platforms.

Here are the disadvantages:

1) The functionality will no longer be there.


Obviously the advantages outweigh the disadvantages if no one is using the 
ray bounder :-) so that is why I am gathering feedback now. Also, if the 
only reason people are using the raybounder is to, for example, render only 
in a sphere-of-interest that functionality can be replaced by something 
other than the ray bounder.


Thanks!

Lisa