So, I’ve used to having the complexity of a surface increase after aligning it curvature to other surfaces, but I must have managed to push from my brain that Rhino doesn’t just do that, but can add entirely new surfaces due to this operation.
I don’t see at all why this is necessary (and yes I know about MergeSrf), but why isn’t there any option to turn this off?
Can you share the file?
Uploaded via the uploader, as this is a live client project.
(Also, already have the workaround by using RebuildUV first, but that has its own issues.)
The surfaces have fully multiple knots. Running the RemoveMultiknot command and checking the option remove fully multiple solves it.
Oof… ok, found out where that came from (Rhino surfaces round tripped through another CAD package).
I wish the MatchSrf command would warn me and present me with that option though, instead of having to go to the forum and asking experts all the time (would probably free you up to do better things). 
Users should be aware that removing multiple knots usually changes the shape of the curve or surface. Frequently the changes are small enough to be insignificant but not always.
I’ll ask what can be done.
yes, that’s true.
Alternatively you could use RemoveKnot once to prevent it from becoming a polysurface, and maybe that is what MatchSrf could offer as an option, since the user is modifying the surface anyways at that end.
How does RemoveKnot prevent it from becoming a polysurface? Also RemoveKnot in general changes the shape. (InsertKnot does not change the shape.)
If I remove one knot, closest to the matched side, the fully multiple becomes a multiple, and no longer results in a split at that point.
Yes, correct, but since the surface is already being modified, a change to the shape is expected anyways.
The desirable and appropriate way to add the option of no multiple knots / polysurface would be to change the portion of the matching algorithm which creates the multiple knots so that it provides a match within tolerances without using multiple knots.
