SubD Bridge: Could work when the ends of the edge sets are touching?

hull.3dm (1.7 MB)
I sometimes have issues with symmetry breaking.

I uploaded a file to illustrate my points. It doesn’t have symmetry problems but imagine that it does.

It’s best to switch off the smooth display to see what I am saying.

Imagine that I found a problem with symmetry in the subd in the original layer.

I would turn off symmetry and delete the side I didn’t want to keep. What’s left is in the halved layer.

But there were some faces that cross the plane of symmetry and it is too difficult to avoid deleting them when I delete half of the model. The missing faces are in the overlap layer.

If I reflect the half model then I get the subd in the reflected layer.

Bridge almost works to recreate the missing faces.

The problem is that it refuses to join the edge loops because they touch at the tip of the overlap region.
I can fix that by creating the single triangle face at the tip of the overlap. Then bridge works.

It would be nice if bridge would just create a triangular face or skip the touching edges and join the rest.

It would also be nice if it stitched touching edges. Edges that were the same length and coincident.

@jmcauley I’m moving your post to a new topic, as I do not think it relates to invalid symmetries in SubD (i.e. SubD object in Rhino that report that they have a symmetry applied, but without the child motif actually being generated).

@Jussi_Aaltonen might be able to help with your Bridge issues.

If I understand correctly, the request is to have this situation be easier to bridge:

Currently, it’s a multi-step process because:

  • The triangle at the top is not accepted by Bridge (Bridge ends need to be disjoint)
  • The subdivided edges at the back won’t weld nicely with the bridge if the number and position of bridge segments is not perfect
  • The selection of edge sets is difficult when there are no corners to stop the valid edge loops (using the Bridge command edge loop selection option, it will select the invalid triangle of edges), and because double-clicking selects edge loops past corners. The Bridge end edges need to be single chain without branches message is not always clear to me.

The SubD is symmetrical de facto, without any Reflect applied, but I don’t think that makes a difference here.

Jussi, any advice here?

DI-173524_Bridge-triangle.3dm (105.8 KB)

My use case might have been a lucky one in that there wasn’t a problem with the edges at the back that we see in your example,

So now I’m not sure how, generally, useful my suggestion is. But it might still have some use with favorable input geometry.

Hi @jmcauley

I filed your requests.

This had been requested before as well:
https://mcneel.myjetbrains.com/youtrack/issue/RH-61737/Bridge-Two-sets-of-SubD-objects-with-common-points-cannot-be-bridged

Added as a new request:
https://mcneel.myjetbrains.com/youtrack/issue/RH-79846/Bridge-stitch-touching-edges-on-sides

  • Jussi