BlendSrf - bad : Loft - good

BlendSrf creates much heavier surfaces between untrimmed edges than necessary to match the edges in most situations. Loft in contrast creates surfaces with the minimum number of control points and spans required to match the edges. This discrepancy is independent of the continuity selected in BlendSrf.

BlendSrf should use the same logic as Loft to create surfaces which are as light as possible while matching the untrimmed edges.

In the examples below BlendSrf creates surfaces with around 60 spans in the transverse direction. Alternative blending surfaces were created by:
Loft between the edges
ChangeDegree the lofted surface in the U direction from degree 1 to degree 5. Retain the degree in the V direction
MatchSrf the lofted surface with the other surfaces.
The resulting surfaces have 7 or 10 control points. The continuity between surfaces is equal to the BlendSrf results.

BlendSrf needs to be updated to work like Loft in determining the number of control points needed.

In the mean time users can create light blending surfaces using Loft > ChangeDegree > MatchSrf

BlendSrfLoft.3dm (1.7 MB)


Indeed, I usually loft etc. if I can. BlendSrf uses Sweep2 under the hood - whatever that does tends to make heavier surfaces.


Based on what I’ve found BlendSrf should be redone using the Loft and MatchSrf technologies instead of Sweep2 technology.

I would love to see more effort in supporting (traditional) nurbs modelling (“sculpting”) with small amounts of CV s and therefor better (post-) editing possibilities. I agree with @davidcockey
Similar to the above example is the missing “simple sweep” since rhino 6.
Just had a student today asking me why the result of a surface looks so complex.
It always takes a lot of steps to find the reason. In this case some controlpoints 0.01 mm of from the knot…

but sometimes those topics do not get much (or no) feedback

1 Like