had my first subd go corrupt today.
image shows a symmetry subd, the left side of image should match but doesn’t. This was working fine for me for over a week on this model, then just this morning effed up on me.
All the mesh points are identical. The creases are identical on each side.
I don’t get it.
Actually not even sure how long its been there as my recent backups have this as well.
Need a mcneel email to send the file to can’t post it here.
I’ve split and resymmetried with no success. Somehow this knot keeps showing up.
Hello- if you can post the object or send to tech@mcneel.com, with a link back to this topic in your comments, I’ll take a look and, no doubt, send it along to the larger brains.
You are correct - it does not work. At least I cannot repeat my initial test which seemed correct but maybe I was hasty. ExtractControlPolygon then ToSubDdoes seem to clean up however… here at least.
-Pascal
@carvecream Removing the crease weight on the edge with the corrupt “bump” (with _RemoveCrease or _SubDCrease to set the weight to 0%), or on its symmetrical parent, will get you back to a normal SubD behavior. Anything else than 0% and the bump is back.
If you move the whole SubD so that this edge is closer to the origin, I think the whole corrupt behavior disappears. You may have to change the weight on the corrupt edge to something else than the current value and back so the display mesh updates.
I just fixed this bug. It was a simple typo in the code where an x coordinate was used in place of a y and z coordinate. Nothing to do with symmetry, distance from origin, or sharpness settings. It occurred on edges with sharpness not equal to 0%, 25%, 50%, 75%, or 100%.
