On Jun 29, 2013, at 11:52 AM, Harmanpreet Singh wrote: >> Excellent, I suggest having your viewer front-end work directly with NURBS >> geometry or facetizations thereof. If you run the brep command on your >> sphere object, you can create a sphere.brep object. The facetize command (I >> think) will turn that sphere.brep into polygons. The bot_dump command will >> export to a variety of standard polygonal text formats you can easily read >> in your viewer. You can use those polygons to display the geometry in your >> viewer. > > 'facetize' command do not run for brep objects. I tried as suggested > by you first by running brep command for an entity say rcc. It created > 'cyl1_brep' object. When run facetize on cyl1_brep, it says: > > nmg_booltree_leaf_tess(cyl1_brep): tessellation failure
Yeah, I was afraid it wasn't yet hooked into the facetize command. The functionality exists, though, so you could make it work as a first step. > But facetize successfully run on implicit primitive cyl1, showing > following message: Facetization is implemented for nearly all objects. It's one of the basic callback functions required for a primitive to be considered complete. However, the method is not sufficiently robust for combination objects (meaning it will often fail). One of the points of NURBS is so we can perform the boolean evaluation in parametric space BEFORE tessellation, so we can get a robust result. Basically, the non-robust current process is: [implicit+CSG] -> [meshes+CSG] -> (boolean evaluation) -> [meshes] -> OpenGL What we're working towards instead that is robust: [implicit+CSG] -> [NURBS+CSG] -> (boolean evaluation) -> [NURBS] -> [meshes] -> OpenGL Cheers! Sean ------------------------------------------------------------------------------ This SF.net email is sponsored by Windows: Build for Windows Store. http://p.sf.net/sfu/windows-dev2dev _______________________________________________ BRL-CAD Developer mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/brlcad-devel
