[ODE] Swept collisions?

[gl]

----- Original Message ----- 
From: "gl" <gl at ntlworld.com>
To: <nlin at nlin.net>
Sent: Wednesday, October 01, 2003 10:48 AM
Subject: Re: [ODE] Swept collisions?


>
> > So we would in this case be trying to collide a sphere sweep volume with
> > a non-swept ragdoll. The best we could do is to take the new position of
> > all bodies in the ragdoll and collide all of them with the sweep volume.
> > However it is still possible for the ragdoll to "tunnel" through the
sweep
> > volume, since the ragdoll itself is not swept.
>
> That's true, but by far the biggest cause of missed intersections in this
> case is the high-velocity object (ie. the bullet), as it jumps massive
> distances in a single timestep.  So using a swept volume for it should
> drastically reduce missed intersections (as long as the targets aren't
also
> moving at high speed - with characters that should be fine).
>
> Of course for a bullet you could just use a short ray and move it along
> every step, but for larger objects a capped cylinder could be useful.
>
> Of course this is the 'accurate simulation vs games' debate again.  This
> isn't completely accurate, but it makes game use much more useful.
> --
> gl
>