Sign in to follow this  
Narf the Mouse

Explain the difference between swept and continuous collision, please

Recommended Posts

Narf the Mouse    322
I'm fuzzy on the strengths and weaknesses of the two methods, as well as the differences and exact details. My main interest in this is physics.

My understanding:

Swept collision: The bounding models are mathematically extended, using a current and last position for each and then a "standard" bounding test is done. Swept collision "sweeps" the models along the path they took.

Continuous collision: The bounding models are combined with velocity and mathematical tests are done relating velocity, position and the models. Continuous collision mathematically tests along the entire route covered by the given velocity.

Why might someone use one or the other, especially for physics?

Thanks.

Share this post


Link to post
Share on other sites
jyk    2094
I'm not aware of a formal distinction between the terms 'swept collision detection' and 'continuous collision detection'. (There may very well be such a distinction, but if so, at the moment at least I can't remember ever hearing it mentioned.)

In any case, it sounds like you're asking about continuous tests that return a boolean result only vs. continuous tests that return information about the collision such as time of impact, collision point and normal, etc.

An obvious advantage of the latter is that information about the collision is usually needed in order to resolve it in a reasonable way; if all you know is *whether* two objects intersected over the time step (but not where or when), your options are somewhat limited in terms of what action to take.

Also, simply testing the extruded shapes against each other may return false positives, which may be undesirable in some circumstances.

Share this post


Link to post
Share on other sites
Narf the Mouse    322
Quote:
Original post by jyk
I'm not aware of a formal distinction between the terms 'swept collision detection' and 'continuous collision detection'. (There may very well be such a distinction, but if so, at the moment at least I can't remember ever hearing it mentioned.)

In any case, it sounds like you're asking about continuous tests that return a boolean result only vs. continuous tests that return information about the collision such as time of impact, collision point and normal, etc.

An obvious advantage of the latter is that information about the collision is usually needed in order to resolve it in a reasonable way; if all you know is *whether* two objects intersected over the time step (but not where or when), your options are somewhat limited in terms of what action to take.

Also, simply testing the extruded shapes against each other may return false positives, which may be undesirable in some circumstances.

Ah - I'd not seen a formal definition of either, so I was going by "best guess". If they're different terms for the same thing, that would explain it.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this