ToSubd fails to create

Hello, I’m playing around with the new subd tools for a bit and found an input mesh for ToSubd which creates a valid not bad SubD, which does not get drawn:

The purple mesh is the input, the subd in question is on the SubD layer.

Maybe somebody knows whats going on here? Feels like a bug…

Rhino_SubD_Bug.3dm (40.0 KB)

Hi Lando,

There is still work to be done to be able to deal with SubD’s with non-manifold edges. It’s a known issue and as far as I know, not on the list to be fixed for the Rhino 7.0 release but will be in a later 7.x release. I looked briefly in the YouTrack database and didn’t immediately find the item but will look some more and add this file to it.
Thanks,
-wim

I may be wrong on this one, but this mesh is not something I would expect to work as a SubD model. It’s all triangles with non-manifold edges.

Most SubD modelers will tell you to model all quads and avoid non-manifold edges. Smoothing triangles and N-Gons pretty much never gives a good result.

I did not expect this input mesh to work, it was more of a curiosity, testing the limits of the subd code.

What is wierd to me, is that Rhino creates a subD from the input which passes the BadObject test but does not get drawn. I would much prefer for the release version of R7 to not create a bad subD from bad input because for a user this just leads to wierd results:

  • The subD gets created but there is no visual clue that it is now present in the model
  • SelSubd now selects the invisible subd but shows no gumball
  • PointsOn shows the control cage of the input mesh and the grips are drawn incorrectly during transformations

Agreed Lando, thanks for the report. I filed RH-56187 to prevent ToSubD from accepting non-manifold edge input for the time being.

1 Like