[ODE] dCylinder problems and future

[Amund Børsand]

I can't help but to throw my opinions on you. :) Why would anyone want
to use such a physically correct engine for something as non-physical as
to "simulate" a person with a cylinder? ;) Just teasing. I'm using it
for vehicles, and all my years of car-sims I feel that wheels rotating
around the Z axis is "logical". However, which axis is which is of very
little importance, I'm using transform geoms anyway (which by the way
are very clumsy to set up and use I think, and I would like if ordinary
geoms could have the "transform" feature built-in, but I don't know the
internals here, maybe it's not a good solution) the main point is that
the CCyl and the Cyl are facing the same direction.

Just my three cents.

- Amund

"John Miles" <jmiles at pop.net> skreiv:

> Actually, I'd rather see the CCylinder rotated so that it faces the same way
> as CylinderClassUser.
> 
> Two huge assumptions: (1) The main application for a flat-ended cylinder
> bounding volume seems to be characters, because they don't tip over and they
> can always turn in place.  (2) Upright characters in most game worlds that
> adopt Max/Maya-like frames of reference are aligned with the Y axis.  Ergo,
> CylinderClassUser is "right" and CCylinder has it wrong. :)
> 
> I'd further assume that CCylinder's inertia-tensor function would be fine
> for CylinderClassUser as well, if the axes were switched.
> 
> -- jm
> 
> >
> > That would be fine, but it should be rotated so that it faces the same
> > way as the capped cylinder. Also, a mass function for it would be nice.
> >
> >
> 
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode

-- 
Amund Børsand <amund £ offroad . no>