Connect a pattern over 2 surfaces / equalize surface parametrisation

Dear Grasshoppers

I've been trying to connect a pattern over two surfaces that share a common edge.

The surfaces are created by curve networks.

I don't get the parametrisation of both surfaces similar enough to connect the pattern on those surfaces at their shared edge.

It should be a simple task to get matching patterns but I haven't found anything that works reliably descent. I have had this problem already with a number of projects and most of the times I ended up repairing the result by hand. You would help me a great deal solving this problem:)

Thanks in advance.

Ps: Using the great Panelling Tools with components like “ptSrfParam” that ignore the surface parametrisation of the inputs didn't work. (see the attached file) Rebuilding the curves or the surfaces in Rhino or Grasshopper didn't work either.

Attachments:

connect_patterns.gh, 24 KB

Replies to This Discussion

Permalink Reply by Richard Schaffranek on September 26, 2017 at 2:57am

Try a mesh and apply supdivisions e.g. Catmull clark subdivision + Diagonlisation from Kangoroo...

Permalink Reply by Daniel Piker on September 26, 2017 at 3:13am

Like Richard says - something like this:

Attachments:

matchmesh.gh, 19 KB

Permalink Reply by Arwed Gollner on September 26, 2017 at 8:37am

Thanks Richard and Daniel.

Pretty neat

Permalink Reply by Arwed Gollner on September 26, 2017 at 6:16am

Hi Tom.

Thanks for the reply. I am using technique 2 most of the times ( I was somewhat aware of technique 1 but never used it, the third sounds interesting, I'll try that)

But problem wasn't to get a good base surface to put my patterns on. Its more that in some cases I need a hard edge between 2 surfaces in order to follow a specifique line with the pattern instead of one big surface that covers all.

The result is very often what you see in my picture. The applied gridpoints/patterns don't match at the shared surface edge due to slightly different parametrisation along the base surfaces.

But I just came up with something that solved it, I guess.

Check out:

Attachments: