[ODE] Holes

Antoine Beyeler abeyeler at dplanet.ch
Wed Mar 21 07:51:53 MST 2007


Hi All,

If your cylinder has a radius of (HoleRadius-BallRadius), then as  
soon as you have a contact you know your ball is on the very edge of  
the hole (and will, in reality, fall). This sounds like the best  
approach to me (although I'm very unexperienced as well). Such an  
approach would avoid the need to consider the depth of collision  
which sounds scary to me ;-)

Best,
Antoine

On 21 mars 07, at 11:14, Christoph Frick wrote:

> On Tue, Mar 20, 2007 at 08:33:19PM +0100, naspiri wrote:
>
>> My first mail to this list, glad to be here :-)
>
> so this is my first attempt in helping out here - so both will be
> flamed ;)
>
>> ... My english isn't so well, I only hope it's understandable...  I'm
>> a newbie, using ODE in a pool game, I made a demo with balls rolling
>> and colliding on a table, but now I would introduce the holes... How
>> can I make that? How can I create a hole on a plane with ODE?  I  
>> think
>> there are different ways and one is trimesh I suppose... is it the
>> only option?
>
> one part i really dont like about aproximating holes with a trimesh is
> the fact, that there might odd things happening (speaking out of pure
> inexperience) - depending on how much tris you are acutally using.
>
> for a pool simulation i would consider experimenting in the area of
> using a cylinder where the hole is (this way you can generate  
> different
> table sizes (7", 8", 9", 12") and also ball sizes programmatically
> (regular, snooker) - and snooker tables have round edges near the
> holes).
>
> when colliding and the sphere with the cylinder and the depth of the
> contact is bigger than the radius the ball falls in. does this sound
> reasonable to the ode gurus here?
>
> of course this way you would not be able to account for effects like
> balls jumping out of pockets again because the ball would not really
> have a hole to fall in - just a "place" on the plane where it happens.
>
> -- 
> cu
> _______________________________________________
> ODE mailing list
> ODE at ode.org
> http://mooshika.org/mailman/listinfo/ode



More information about the ODE mailing list