Rebuilding Daniel Pikers courtain simulation

Hey guys,

I tried to reconstruct Daniel Pikers courtain simulation.
As you can maybe see in his simulation, all scaled points are target points, and so the cloth sticks to all of them.
However, my scaled upper point row does not get registered as target points for anchoring. Only the last and the first from the row.
I tried a bit around with some force multiplicators, but it still doesn’t work right.

Also nudging does not work really, allthough I think it could be because of some force issues.

Some help would be apreciated alot…thanks in andvance…I have my presentation next (33.9 KB)

Are you starting from the defintion posted here?:

I was missing some plugins/internalised data in your definition, so couldn’t run it, but it sounds like it is maybe just that you are using points for anchors which do not coincide with vertices of your starting mesh.



If manually choosing which points are the ‘hooks’ of your curtain to anchor, you can bake the initial mesh into Rhino and make sure you choose points from those vertices.

1 Like

Hey thanks allot first.

so the best thing would be using the initial verteces, their copys, or closest verteces to my chosen points?

Why would that make a difference to randomly chosen points? I think thats the part of the definition where I lack the knowledge.

Hi Gabriel,


Inside the simulation, Kangaroo always works on points (or oriented points if using the 6dof stuff).
Goals act on combinations of points, and geometry like this curtain mesh is defined by the positions of its points.
Since the mesh is essentially just a collection of 1d springs, if another goal, such as an anchor point, acts on some point which happens to lie on a face of this mesh, Kangaroo doesn’t know that you want it to act on the mesh. Only if it coincides with a vertex.

that makes sense.
Tanks allot for this information!

You saved my presentation.

Im trying to achieve something similar like this…

would give me headaches modelling all this…