RBD pin constraint network
1991 3 1- AndyW
- Member
- 336 posts
- Joined: Dec. 2014
- Offline
I'm new to constraint networks in Houdini and this one has me a little stumped. In my example file I need to connect the array of spheres to each other vertically and anchor them at the top like a bead curtain but as you can see, setting the Search Radius on the connect adjacent pieces gives me horizontal connections instead of the pieces above and below. I'm also not clear on how I would anchor the whole chain of spheres at the top. Ideally this setup is fully procedural, so that size and number of beads can be changed without building new constraints. Tips are greatly appreciated!
- ludwigvon
- Member
- 27 posts
- Joined: June 2016
- Offline
- matthias_k
- Member
- 483 posts
- Joined: Dec. 2006
- Offline
pointlist is your friend, too
create a dynamic group in your geo,
then use:
pointlist(“pathtothesop/groupname/”,“groupname”)
have in mind: manual says:
pointlist(surface_node , group_name )
but if you have a scene like yours, and you try:
pointlist(“…/beads/”,“groupname”)
pointlist will search for the “groupname” in “OUT”
where the groupname is maybe no more existent because of the dopimport.
If you do not need eaxct collision on the spheres, then it's maybe much
faster to use wires, set the collision radius of the wire to the sphere
radius, then copy the spheres on the final simulated wire.
Self collision added… then you can see, that the “dark” side
are intersections can happen in the wire collision.
Maybe width adjustment can fix it. But: it's fast…
create a dynamic group in your geo,
then use:
pointlist(“pathtothesop/groupname/”,“groupname”)
have in mind: manual says:
pointlist(surface_node , group_name )
but if you have a scene like yours, and you try:
pointlist(“…/beads/”,“groupname”)
pointlist will search for the “groupname” in “OUT”
where the groupname is maybe no more existent because of the dopimport.
If you do not need eaxct collision on the spheres, then it's maybe much
faster to use wires, set the collision radius of the wire to the sphere
radius, then copy the spheres on the final simulated wire.
Self collision added… then you can see, that the “dark” side
are intersections can happen in the wire collision.
Maybe width adjustment can fix it. But: it's fast…
Edited by matthias_k - Jan. 7, 2017 10:54:00
English is not my native language, sorry in advance for any misunderstanding :-)
- AndyW
- Member
- 336 posts
- Joined: Dec. 2014
- Offline
-
- Quick Links