"Christoph Hormann" <chr### [at] gmxde> wrote in message
news:3BFF5FDD.DC431EA0@gmx.de...
>
>

> >
> > I've modified my particle interaction routine in which particles
experiment
> > a cohesive force, but cannot be too close either, so as to avoid having
two
> > distinct particles occupying the same space.
> >
> > This can be seen at the end of the sequence, in which the blobs settle
down
> > in the bottom of the "receptacle" and don't occupy the same space. This
> > really improves the sense of the conservation of volume, which was
evident
> > in some past animations.
>
> But there still seems to be a problem with the energy, the particles are
> 'vibrating' quite a lot in the end.  I suppose this is a general problem
> of particle based simulations since they somehow behave like molecules and
> this is exactly what the single molecules are doing in a liquid.

Yes, the way it is simulated causes that vibration. I don't like it either;
maybe a finer simulation will help reducing the magnitude of the vibration
and make it less evident.

> > I'll be happy to receive any feedback on this sample.
> >
> > Of course, I'm also fed up with this surface, so the next thing I'll try
is
> > to extend the algorithm to accept non-lathe surfaces.
> >
>
> That's the advantage of doing things in SDL or as a Povray-patch.  You
> have access to the internal intersection routines.

One of the advantages of my algorithm is that it can be parallelized, that
is, I can use several machines to code several frames independently, because
the coordinates of the particles are given by my C program. Can it be done
similarly with SDL or POVRay-patch?

Thanks for the suggestions,

Fernando.

Post a reply to this message