Testing Point Inside Mesh Node

[enzyme69]

point_inside_002_2017_07_24_13_03.zip

I am testing this node, seems like pretty intuitive, although I might be doing it wrong?

The idea is to somewhat fit the box elements perfectly, kind of creating Remesh, but each voxel box fits inside the 3D mesh input.

[enzyme69]

ansv_points_inside_outside_001_2017_07_24_13_35.zip

AN BVH Inside thing seems to give more accurate points inside.

Wonder why...

[zeffii]

there's a thread for this already.. https://github.com/nortikin/sverchok/issues/1740 and a branch https://github.com/nortikin/sverchok/tree/pinside_mod

AN BVH Inside thing seems to give more accurate points inside. Wonder why...

different code, different algorithm

[zeffii]

These kinds of algorithms should be implemented as core Blender api calls.. so all add-on devs / scripters can use them, and they would be faster.

[zeffii]

look at AnimationNodes' implementation of such algorithm: https://github.com/JacquesLucke/animation_nodes/blob/c44280aae363402797e5da6319e755861b800380/animation_nodes/nodes/bvh_tree/is_inside_volume.py

read that and interpret what it does for yourself.

What do I think?

I think the multiple random direction raycasting from each test point might be the strength of the algorithm. I'm happy to add variations to the node. Excellent project for someone else.

[Durman]

@enzyme69 Points inside mesh have normal working. Do not use mask socket, the points out is already masked. Or use mask with primary points. I a little modernized your node tree.)) https://gist.github.com/f3be5fe21fb571b2fa7797c4b8e50add

[Durman]

@nortikin Constraints rotation is good, thanks.)