[ODE] static Meshes

[Megan Fox]

Well of course you'd implement the stairs as box geoms too ;)  As far as I
know, everything that can be represented by a solid geom should be
represented by a solid geom, with trimesh being reserved for those cases
that can't be described by a collection of default solid geoms.  Tri-mesh
can be used if necessary, of course, but it seems to lead to an increased
chance of objects warping through eachother, or other instability.

That said, this approach really works better if you're already building your
stages out of component chunks.  If you're doing BSP style big-mesh levels,
I suppose it could be quite time consuming to represent it all via solid
geoms (or maybe people still do it that way anyways, no idea).

-Megan Fox

> > In your case, a chair can simply be a box. Or two boxes - the one that
> > bounds chair legs, and the other one that bounds backrest.
> Simple and fast.
>
> That's strange. I heard the initial request as "a chair falling
> down some stairs".
>
> The absolutely obvious way of modelling this is to turn the
> stairs into a series of boxes, and use a mesh of the chair. This
> also gives good locality, as you won't test the entire stairs
> against the chair each frame (bounding volume test will discard
> most steps). Then, create the inertia tensor of the chair by
> adding a bunch of box bodies (not geoms!) for the legs, seat and back.
>
> Cheers,
>
> 			/ h+
>
>
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode
>
>