generative modeling for Rhino
I posted this in the general discussion, guess that was wrong... so here we go again:
You can see in the picture I have a certain amount of curves along which there are some points distributed. The curve end points are fixed, but the ones inbetween are supposed to distribute themselves in order to be as far away from their neighbours as possible (the blue dots). In other words, all the points should be distributed evenly while still remaining on their respective curve.
What the Kangaroo definition does so far is pushing all the points towards the curve ends as the curves are open and therefore the general force is of course towards the outside.
Instead, the points should keep their approx distance towards the neighbours on their curves and just move so that they are placed inbetween the neighbours on other curves.
not quite, especially with several curves the hole setup seems to be undecided. The points either move towards the curves' ends (repulsion is stronger) or they just ignore the repulsion and distribute equally (spring force is stronger).
The problem seems to be the way Kangaroo is set up as the connections between the points isnt updated. Otherwise one would only need to define the closest point connection which would then change depending on the point's position.
So I tried some other things yesterday and HoopSnake seems to work pretty well. However, now the actual physical simulation embedded in Kangaroo is missing, of course. With a few tricks a colleague and I managed to combine HS with Kangaroo in order to feed Kangaroo's output into the initial input again, which opens up many new applications (!), but also slows the whole thing down quite considerably. We wanted to make a seperate post about this actually, but that's a different topic.
So, in conclusion, it's not quite working yet the way I thought it should be, but I'll find a workaround.