[ODE] Adding Negative Masses

[Auro]

Of course ur right, (though I'd still wonder what wud happen to the inertia
tensor after the -ve addition, wud it still remain +ve definite? if yes how
do u show that?) Also my point is irrelevant if the final total mass is
positive, then of course ur expecting everything to be fine excepting
numerical errors. What I meant was in the case of a -ve mass itself. That
definitely is contradictory to the assumptions of physics - I mean certain
equations wud probably still be solved fine, but u'd get funny results --
like u mentioned friction speeding a body up rather than slowing it down.
But using this strategy for simulating a balloon? That's more like assuming
a negative value for a distance (norm).
 -Auro

On 4/26/06, Jaap Stolk <jwstolk at gmail.com> wrote:
>
> On 4/26/06, Auro <auro.ghosh at gmail.com> wrote:
> > Negative mass? wouldn't that mean that the Inertia-Tensor would probably
> > become negative-definite (or atleast wouldn't any more remain
> > positive-definite). Shouldn't that bug certain laws of physics? Just
> > wondering... would be nice to have someone pop in the proofs :).
> >  Cheers,
> >   -Auro
>
> I think the idea was to remove part of the mass using a negative
> dMassAdd().
> The resulting mass will still be positive, so I think physics should be
> ok.
> There may be a problem with subtracting the last mass, since all that
> remains are (accumulated) rounding errors, which can be anything.
>
> If you go as far as actually ending up with a body with a negative
> mass, that may be a problem. friction may cause it to accelerate etc.
> :-)
>
> It's simple to test. add two masses together, then add the last one
> but with a negative mass, then look at the mass properties. should be
> the same as the mass you started from.
>
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode
>



--
-----------------------------
- Aurobrata Ghosh -
+91-9873159692
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://q12.org/pipermail/ode/attachments/20060427/39622f6d/attachment.htm