[Sidefx-houdini-list] RBDs - Collisions with lots passive moving pieces, most efficient way ?

Rafael Santos rafaels at gmail.com
Thu Jan 19 12:21:45 EST 2012

If all the rocks have similar LOD, I'd say your best bet is to create an RBD Glue object and set it's "create active object"  toggle to off.
That'll give you one SDF per rock but each one will have the max detail level you specify in the collision volume page, rather than the bounding box of the merged rocks.

Now if you have rocks with different LOD you might want to create sets base on their LOD and create different RBD Glue objects so you can set the collision volume resolution according to the LOD of the different sets of rocks.

Ultimately if every single rock is a different beast, you have no choice rather than creating different RBD static objects... :P
I did once have an issue with creating one single SDF for complex groups of geometry where it created pockets of concavity and the collision detection failed in those cases... But that was a long time ago... in this very galaxy...


Rafael Santos
rafaels at gmail.com
[416] 897-2841
Toronto - ON

> Hi,
> I have to make little rocks collide with quite a bunch of keyframed
> geos (around 100). (translating/rotating objects, not deforming)
> What you would say is the way to go for this :
> - Generate one SDF per object, apply to these SDF the key framed
> animation and use that in DOPs for collision with the proxy volume on
> the collisions/volume tab .
> But this means 100+ hundreds different static objects in DOPs
> - Generate one huge SDF (takes quite some time with a sufficient
> resolution to get proper detail) per frame and having just one static
> object in DOPs, again using the proxy volume.
> Sadly I don't really have time to make proper performance tests right now.
> Thanks a lot for tips and pointers,
> Vincent
> ------------------------------
> _______________________________________________
> Sidefx-houdini-list mailing list
> Sidefx-houdini-list at sidefx.com
> https://lists.sidefx.com:443/mailman/listinfo/sidefx-houdini-list
> End of Sidefx-houdini-list Digest, Vol 330, Issue 2
> ***************************************************

More information about the Sidefx-houdini-list mailing list