Better blend surfaces

[samuelmattias]

This is how I make surface blends in Rhino:

1. Duplicate edge curves

2. Turn on edit points
   

   

3. Extract isocurves at edit points locations
   

   

4. Blend curves
   

   

5. Make a 2 Rail Sweep
   

   

This results in a nice single-span, up to G4 continuous blend surface and relates to the GH Blend Surface definition post I wrote previously.
I might be wrong, but it seems to me that this method produces better results than making the usual 2 rail sweep and then running a surface match command.
Indeed, matching the surface breaks G2 continuity at the naked edges and even at the internal cross sections end points, even if I set the ‘Isocurve Direction Adjustment’ to ‘Automatic’ in the Match Surface options window.

So my questions are:
Shouldn’t the match command try to achieve a ‘natural’ isocurve direction when set to automatic?
It seems intuitive that two surfaces have G2 continuity at one edge if their isocurves are G2 continuous, am I wrong?

[pascal (Pascal Golay)]

Hi Samuel -

So far I do not get the edge falling away as you show - if both of the original surfaces are untrimmed (?) along that edge then MatchSrf > Automatic should line things up nicely - can you post your file with all the curves etc?

thanks,

-Pascal

[samuelmattias]

Hi Pascal, this is the file:
blend.3dm (308.3 KB)
I checked again but the misalignment still occurs. Regarding the match options, I meant ‘Match target isocurve direction’ instead of ‘Automatic’, sorry.

[pascal (Pascal Golay)]

Hi Samuel - I see - thanks. I think I’d expect little or no change at all in the surface. I’ll see what I can find out.

-Pascal