ICE collision "event"...?

[Thomas Helzle]

Hm, I seem to be a bit out of the ICE loop...

I tried for quite a while now to simply switch a particles states if an ICE rigid body collision between particles or an obstacle occurs but I can't for the life of me find anything useful in the docs or on this list. How can I get a "collision event"?

Thank you very much!

Thomas Helzle

[Steven Caron]

in physx these are call 'contact reports' i believe... they aren't available in ICE.

s

[Thomas Helzle]

Does that mean that I can't find out if and when a collision occurred?

Please tell me this isn't true :-(

Thanks

Thomas

[Steven Caron]

not with the simulate rigid body node natively. you would need to use the basic collide node and derive it yourself. unfortunately if you are using instanced shaped rbds, you wont be able to get an accurate collision using the basic collide node. the basic collide is only going to work with other shape types than instance.

s

[Thomas Helzle]

Wow - I'm impressed - on the negative scale.

I would have thought this would be the most basic thing to do... :-(

Thank you very much, so I guess the Rigid Bodies in ICE are still pretty much unusable.

Too bad.

Cheers,

Thomas

[Andrew Nicholas]

Yep, until we get actual shape collisions in there too, I'd partially agree with that diagnosis. I'm looking at Bullet in Houdini at the moment. It's got a fair way to go in terms of features but at least it does proper collision. Not as fast as ICE though, even with bounding boxes.

[Thomas Helzle]

RBDs in ICE feel like a "demo feature"  -  looks nice on stage but seems rather useless in reality.

Especially since RBDs combined with the power of ICE could make them _really_ killer...

Too bad,

Thomas

[Andy Moorer]

Pretty much. You can use them for a lot of stuff, particularly "supporting" other effects, but that said I am still going outside ICE for most of my RBD work, for the very reasons mentioned here. 

I would put Improvements to RBDs in ICE as #2 on my wish list, right after support for creating/modifying geometry.

(#3 is some decent support for realtime shading.)

[John Pierre van der linden]

I agree with andy but i think the order is different, but i think i'd put it at #1 as create/modify can be more easily done externally. in/out of rbds is a lot harder.  Then #2 realtime shading, + ability to have the hardware render implement AA etc, as much could be done this way. FYI yes they are called 'contact reports' and on gamedev you can see code for it. Still prefer physx but looking at bullet like andy to get away from these + other problems.
JP