Hi,
I've ran into this problem in several of my setups. I'm starting to think this is a bug rather than me just being ignorant. Can anyone confirm?
Quite often I get these odd bits left hanging in the air glued to the “parent” static rbd object when they clearly should fall down freely. I've tried with the RBD Glue Object and the Glue Adjacent methods with similar results. Have a look at the attachments.
Any ideas or workarounds?
Cheers,
Kari
Buggy behavior RBD Glue Object and Glue Adjacent
6835 4 0- kbrown
- Member
- 4 posts
- Joined: Oct. 2012
- Offline
- KiLa
- Member
- 109 posts
- Joined: June 2008
- Offline
- kbrown
- Member
- 4 posts
- Joined: Oct. 2012
- Offline
- essencevfx
- Member
- 19 posts
- Joined: April 2009
- Offline
- derrick
- Staff
- 329 posts
- Joined: July 2005
- Offline
kbrown
Quite often I get these odd bits left hanging in the air glued to the “parent” static rbd object when they clearly should fall down freely. I've tried with the RBD Glue Object and the Glue Adjacent methods with similar results.
There are two different types of glue system understood by the Bullet solver: RBD-style glue and the Glue Network Constraint-style glue. The two glue systems have different behaviour. With the RBD-style glue, a piece needs to be directly involved in a collision to be released from the glue parent (this is why some pieces remain glued to “box_passive”). With the Glue Network Constraint-style glue, glue bonds further away from the location of the collision can be weaked and release (see the “Impact Propagations” and “Propagation Rate” parameters).
kbrown
Any ideas or workarounds?
Consider replacing the RBD-style glue in your simulation with Glue Network Constraint-style glue by clearing the “Glue to Object” parameter on the /obj/AutoDopNetwork/box_active node and adding “box_passive” to the “Object Mask” parameter on the /obj/glue_piece1_to_piece9/fetch_objects_to_glue node.
-
- Quick Links