PDA

View Full Version : VPR using Render SubPatch Level instead of Display SubPatch Level



rsfd
08-19-2011, 10:05 AM
VPR is currently (10.1 #2161) using the object's Display SubPatch Level.
Being a Preview Renderer, I would expect it to use the object's Render SubPatch Level instead.
Any reason for the actual status quo?
Pros and cons?

Lightwolf
08-19-2011, 10:27 AM
Yes, the reason is that the display subpatch level is the geometry that's available while working interactively, while the render level require freezing and also kicks out the mesh created for the display.
Essentially it's a matter of speed - and LW also being able to store the geometry created by either of them only... but not by both of them at the same time.

Cheers,
Mike

rsfd
08-19-2011, 12:26 PM
Thanks for the head ups, Mike!
Makes sense (so I won't file a FR on that ;) )

evolross
08-19-2011, 12:30 PM
There's mixed feelings about this because a lot of people argue that for heavy deformation work (e.g. head models that use Z-brush deformation and normal maps) having VPR only work with Display Subpatches will not allow them to increase the subdivisions enough to ever see any detail because Layout would be too slow because of polygon limitations with OpenGL. And in it's current state, VPR would probably become too slow to be useful with that many polygons anyway.

UnCommonGrafx
08-19-2011, 02:46 PM
that's where the gigapoly mesh handling capabilities will need to be available to us.

Is that trademarked? "Gigapoly"

Core, in theory, had the nugget to do this. Should be coming. ?