Serengeti has been what I would call “aggressively” splitting my surfaces apart at different times, despite turning off that feature in advanced settings.
It seems to happen when filleting certain surfaces. Note that when I run RemoveMultiKnot on the upper surface, I get a single fillet and no broken apart surface, but I haven’t had to do that in previous Rhino versions.
Surface_split.3dm (2.4 MB)
@menno any ideas here?
The command has a trim option. Here I unchecked it unfortunately that does not trim the surface…
But _FitSurface helped…
@Martin I’m not sure what your deal is, but I’d appreciate it if you’d ignore my posts from now on please.
I’m not seeing martin be anything but helpful here…Maybe you are misunderstanding his intent here?
we are all friends here trying to help each other…
Thank you for the reminder.
@Gijs @menno When current V9 WIP FilletSrf with the Trim option is used on a surface containing multiknots that surface is converted to a polysurface (split at the multiknots). It does not occur without the Trim option. It does not occur with V8.
Example: Blue surface has multiknots created by changing degree from 2 to 3.
FilletSrfMultiKnot.3dm (2.2 MB)
yeah this feels buggy-
this feels buggy for sure-
RH-92139 filletsrf tim= yes splits surfaces to polysurface
I agree, the Trim option should behave like the Trim command. The command bypasses the setting of CreaseSplitting and returns a trimmed fully-multiple knot result, and not a polysurface. FilletSrf and related commands should do the same.
For now, as a workaround until this is fixed, you can (temporarily) disable this by issuing _CreaseSplitting _Disable prior to using FilletSrf and restore it with _CreaseSplitting _Enable. It is generally a good idea to keep crease splitting enabled.
@menno It is interesting that CreaseSplitting is affecting a surface without fully multiple knots. The surface in my example above does not have any creases (no fully multiple knots). It is a degree 3 surface and the knots have a multiplicity of 2, not 3 as needed for tangential discontinuity in a degree 3 surface.
I see that, making this even more curious. I’ll report back when I know more.
Update: this is specific to FilletSrf and ChamferSrf, and clearly needs to be tuned up. I will get to this before v9 is released.
