Molecular orbital rendering with Flying Edges is incorrect (negative mesh)

[ghutchis]

Here's a rendering of the LUMO for benzene (file is attached - unzip to get benzene.fchk):

With the current code, I get a very different look:

The problem is with the mesh for the negative isosurface (it's the mesh2 in the code). The code sets m_reverseWinding and inverts the iso value (i.e., not 0.03 but -0.03)

For example, the triangles are a bit weird too:

benzene.fchk.zip

[ghutchis]

@perminder-17 - any thoughts on this? It seems as if the Flying Edges is gluing together the two parts of the mesh even though the isovalues are very small.

[perminder-17]

Hmm..probably the issue behind is with m_reverseWinding. I'll check it very soon, in marching cubes algorithm we sets the vertices and normals by looking at the m_reverseWinding but in flying-edges I think I have missed that. I'll fix it very soon, Sorry for the oversight.

[ghutchis]

No worries, I should have tested more carefully. I'm just working towards a release in mid-December, so everything is getting tested more rigorously.