Depth buffer and "scene" building

Some of my initial ideas for vvvv are somewhat “scene” oriented, calling for a large geometry such as a sphere that other elements will be place inside. But I must be missing something thus far. The attached scene uses the SoftImage camera for convenience. I can’t seem to get “inside” the half-sphere. If I navigate to where the sphere is open toward the viewer, zoom as far as possible, then rotate the view left right, I still get the outside of the sphere showing up once I reach the halfway point. I have the depthbuffer format set to D16_LOCKABLE. Any ideas on getting inside the sphere? Am I perhaps hitting some collision detection built into the softimage camera patch?

spherescene.zip (59.0 kB)

hi cwiggins,

instead of zooming the camera in (z key), you should try to dolly closer (p key). zooming does not change the position of the camera, so it remains outside the sphere. dolly it inside the sphere, and your view will no longer be obstructed by the outside of the sphere.

you can also use a Cull node to discard certain faces of the mesh alltogether, but i believe this won’t be necessary in your case.