Collision volume of complex static objects

   2488   2   0
User Avatar
Member
56 posts
Joined: July 2017
Offline
Hello-
This idea started in another thread on a pinball machine simulation, but the question has evolved into one of the collision volume of complex static objects. To illustrate the question in simple terms I have created a HIP file which consists of a sphere which is caught in a “bumper,” which is simply two boxes arranged in the shape of a “V.”

The sphere is an RBD hero object, and the bumper is a static object. I would have expected the sphere to fall into the interior of the “V,” and get caught there. Instead the sphere lands on an invisible plane connecting the two sides of the V. I assume it is an issue with a parameter in the dynamics, but I am having trouble locating the parameter.
-Scott

Image Not Found


Edit: Hey, I figured it out. I just had to change the solver from bullet to RBD.

The thing is, the original pinball simulation doesn't run as smoothly, as if the game board and/or ball have very bumpy surfaces. I'm fiddling with the friction…
Edited by scottval - April 12, 2019 19:21:34

Attachments:
Bumpertest.hipnc (340.5 KB)

User Avatar
Member
9421 posts
Joined: July 2007
Offline
don't use old RBD, it is unnecessarily slow and much less flexible than bullet
just check Collision/Bullet Data/Create Convex Hull Per Set Of Connected Primitives on your “bumpers” DOP

also for bullet use only Bullet Data tab to modify and visualize collisions, any settings and visualisations on RBD Soler tab are ignored by bullet
Tomas Slancik
CG Supervisor
Framestore, NY
User Avatar
Member
56 posts
Joined: July 2017
Offline
Yea, I took your advice re. bullet/convex/per set of connected primitives, etc., and it works great, thanks.
  • Quick Links