generative modeling for Rhino
Actually any idea about how to locally change/deform a grid density while keeping the edges of the grid as they are and the grid flat would be helpful :)
A quick anwser could be use a packing circle box ( like this http://www.grasshopper3d.com/forum/topics/packing-circles-on-curved), and then put inside the hexagon.
The other option, (that I think you are looking for) would be starting from a very little and constant grid of triangles, and then changing the number of points that build the hexagon, depending on the distance of the point
To conect the circles by lines, is easy using "delanauy edges" (in mesh components), mades the more natural grid. The problem is to group the lines into hexagons, I don`t know now how is possible to done....
Muchas Gracias Manu !
I have no idea how to organize delauney edges into a hexagonal pattern.
This might deserve a question on its own!
If your Delaunay triangulation has 6 triangles around each vertex, then the Voronoi decomposition of the vertices will be made up of hexagons.
Actually the points do always have 6 triangles around each vertex :)
The hexagons are really hardly recognisable though:
I have found a cool definition from Co-de-it doing what I wanted, but when using several attractors there is a weird separation between the cells, not sure how to fix that:
I also found a great applet by n-e-r-v-o-u-s which uses springs from "traer.physics" to stretch the hexagonal grid, would you know if the same is possible with the Kangaroo's spring component?
Would you have a trick to bring these single lines back to hexagonal cells after the Kangaroo component? It'd be great if the lines removed from the "remove duplicate lines" could be brought back. Sounds difficult though.
Rebuilding the cells is no trouble at all - that's what the Geometry Input and Output of Kangaroo is there for. Just connect the vertices from the explode component to the input, and it will be output deformed by the physics simulation but with path and list structure still intact, so you can make closed polyline cells from it