I’ve been toying with some ball/flat surface collision physics for an idea I had, however… It seem as though, even with a continuous detection on both the ball and surfaces and extrapolated results, there’s a speed at which a the ball will pass through a colider.
I haven’t done the math on this yet, but my hunch is that it’s a limit where x is force and y is collider physics, so As x approaches infinity, y approaches a larger infinity. (In that nonsensical math way about how some infinities can be larger than others.
Make the moving object be Continuous Dynamic, and any objects it collides with be Continuous. It will not be able to pass through anything when this is done. Using extrapolation has no effect on collision anyway; physics only runs at the physics timestep no matter what, and extrapolation is purely visual.
There is a speed at which an object will force its way through, but only if youre using Transform movement changes rather than using the rigidbody forces to manage it…
Sounds like youre using Transform.Translate type functions, which will be where you problem is.
Really? I’m sure I’ve tried every physics setting in the book in the past and the only solution was to either increase the scale of my scene, add the DontGoThroughThings script from the wiki, or changed the Fixed Timestep.
Isn’t the real issue collisions don’t fire because between two physics updates the item has come and gone without the colliders intersecting within the physics update?