Hi All, With the osgVolume project my plan was initially just to require Texture3D so we wouldn't have to mess around with stacks of Texture2D as a fallback. Last night on pondering different elements of design it occurred to me that all the rendering techniques of real interest can only be implemented with shaders or on the CPU using race tracing, there simply isn't a way to create a fixed function pipeline. Since the fixed function fallbacks will be very limited compared to the ray traced versions it begs the question whether this limited functionality justifies the extra complexity in providing fallbacks.
Personally I'm rather tempted to conclude that setting minimum spec for osgVolume to be GLSL and Texture3D support (i.e. OpenGL 2.0 era hardware) would be a reasonable thing to do. If you are going to try volume rendering then you need modern hardware anyway just to cope with the demands on memory size and memory bandwidth, and all modern graphics hardware supports at least OpenGL 2.0. Would a minimum spec of OpenGL 2.0 be acceptable for osgVolume users? Thoughts? Robert. _______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
