Reaction Diffusion arlgorithm definition

Replies to This Discussion

Permalink Reply by David Rutten on October 19, 2016 at 4:25am

It looks like it might be reaction-diffusion, but it could also be something more geometric. In fact the strong variation in band width makes me suspect something other than R/D, or maybe R/D with some additional morphing.

The problem with R/D patterns is that they're pixel based, and Grasshopper doesn't have a bitmap data type. So either you need to use a plug-in which does, or use some other approach like meshes with coloured vertices. Either way at the end you still need to convert your sample-space to nurbs curves.

Another thing you can try is to start with a regular pattern (nurbscurves or polylines, perhaps some hand-drawn branching) and then deform it in various ways. You could for example set up a vector field consisting of various forces, then have it affect your curve-control-points. Or maybe move control-points using some sort of equations for X and Y...

Permalink Reply by Michele Segolini on October 19, 2016 at 4:29am

I'll try!thank you very much David!

Permalink Reply by Mateusz Zwierzycki on October 19, 2016 at 5:18am

It looks totally hand drawn, I wouldn't say it's R/D pattern as they usually have a constant width gap... 

This is an example of a similar pattern produced with R/D:

Permalink Reply by Shridhar Mamidalaa on October 19, 2016 at 7:11am

What u asked for looks hand drawn and is not rd pattern.
I tried to do similar effect using absolute values of Perlin noise.

Attachments:

• perl.gh, 6 KB