Access to Physics internal collision

[the-jerry]

Hey,

is there any way to access any of the internal physics collision calculations of TVPhysics? I am asking because, all my collisions are done by the physics engine. I would like to know which two objects collided and checking everything with AdvancedCollision again seems such a vast of ressources. (because TV already has the results)

Thanks for any hints.

[jviper]

NumerOfEvents = TVPhysics.PollEvents() Returns the number of events that occurred
For i as integer=0 to NumberOfEvents-1
    Evnt = corePhysics.GetEventCollisionDesc(i)
Next i
Returns each event with their info.

[SylvainTV]

Yes don't forget to enable events with SetMaterialInteractionEvents though

If you don't want to do physics simulation, you can use the new TestCollision function of TVPhysics that just checks collisions between bodies.

[the-jerry]

Thank you guys. I will try it.

[the-jerry]

Ok, let's come back to this topic. ^^

I tried jviper's code and for now I get no Event back. (NumberOfEvents = 0)
So I guess I must set up the MaterialInteractionEvents right. I am confused by the two material parameters. Can someone explain this a little more detailed please? Why two materials? What do they stand for?

I would appreciate any advise. Thank you.

[sybixsus]

They don't stand for anything. You apply them to your physics objects and then pick which materials should interract with one another, and how.

Example:

Two objects. One is a cannonball, one is a ship - a Spanish Galleon. The ship is (basically) made of wood. So you create a material called Wood and apply it to the ship's body. The cannoball is (basically) Metal so you make a material called Metal and apply it to the cannonball. Now you define how wood and metal should behave when they collide. Done.

[Hypnotron]

define how it should behave?  could you elaborate?  the wood show splinter and break and the cannon ball should be pretty much invulnerable... but it depends on how fast the cannon ball is going.  If the cannon ball is just dropped onto the deck while loading the cannon, the wood shouldn't be damaged (practically speaking).  Can all this be expressed in the settings of the material?

[sybixsus]

Newton does not, AFAIK, support breakable bodies, so no, not unless you did all the hard work yourself. The properties you can define are pretty self explanatory, friction, softness, bounciness, whether they should collide at all and whether they should use continuous collision ( which is to avoid bodies going through other bodies if they're moving too quickly.)