algorithmic modeling for Rhino
This is fantastic! I have been working on a fast way of thickening lines for structural analysis in GSA but have been struggling a lot. Would you be happy to share this definition? Email firstname.lastname@example.org
This definition is similiar to the one on LAN site called gh_shelling (thanks Luis). It should work fine as a 3dprintable version of the original wireframe.
Some notes about this problem - as I noticed, it is very often discussed an has some practical values.
1. Shape, which produces an ideal circle after CC-subdivision is a hexagon, and hexagonal tubes are ideal for long bones, if you have plans to get subdiv.
2. This leads to solving such problem - getting complex joint blending to hexagonal sections, which may be blended without any pain.
3. There may be a lot of additional information about your joints and their shapes, so, I seem, it requires complex approach to each case - you may solve joints by implicit field sampling, but bones - by hexagonal tubes, with smart blending of different-count slices, or you may define template planes (like voronoy-cells walls) for bases, or you may try to build another best solution, which is named "anything else" :)
4. My proposal is next - to start with tube/sphere solution, scatter points at joint area (excluding overlapped surfaces areas) and perform smoothig/relaxing steps with more smoothing, if points has too much different tubes around, and restore/simplify target surface. It is most naturalistic approach :).