What criteria determines where a variable blend surface ends?

Well, since this thread ended up being a general bashing of variable surface blends in Rhino, let me add some fuel to the fire: :wink:

Why is the edge of this blend curved when the iso curves of the original surfaces are straight?

image

Now, this makes me suspect that there’s some weird, underlying problem in the way Rhino interpolates what should be uniform transforms across a surface, because this strange bending also showed up in @Pascal’s skew surface plugin that can be found here.

I’m guessing it’s too much to ask for someone from McNeel with intimate knowledge of the underlying surface math to comment on this?