I wrote the code in C, and now it is much faster! I'm beginning to think
about adding particle interaction. That will slow the algorithm
considerably, but I think it may be worth the try.

I still don't like the appearance of the blobs... it's still too "chunky".
Even though they are 10,000 blob spheres, I can still spot lots of them.
I've tinkered with the threshold and strength parameters, but I haven't
reached any satisfactory values. Any suggestions?

Well, I hope you like it. And of course, your commentaries are welcomed.

Good luck,

Fernando.

Post a reply to this message

Attachments:
Download 'liquid8.m1v.mpg' (262 KB)

> I wrote the code in C, and now it is much faster! I'm beginning to
think
> about adding particle interaction. That will slow the algorithm
> considerably, but I think it may be worth the try.

> I still don't like the appearance of the blobs... it's still too
"chunky".
> Even though they are 10,000 blob spheres, I can still spot lots of
them.
> I've tinkered with the threshold and strength parameters, but I
haven't
> reached any satisfactory values. Any suggestions?

That's much better.  Particle interaction may well be the way to go,
especially adding viscosity (particles sticking together) as this may
cure the problem of individual components being visible.  Also it would
make sure that particles didn't occupy the same space at the same time
(if that is not the case already).  Good work nonetheless.

Post a reply to this message

"Andrew" <ast### [at] hotmailcom> wrote in message
news:3bfd5a15$1@news.povray.org...
> > I wrote the code in C, and now it is much faster! I'm beginning to
> think
> > about adding particle interaction. That will slow the algorithm
> > considerably, but I think it may be worth the try.
>
> > I still don't like the appearance of the blobs... it's still too
> "chunky".
> > Even though they are 10,000 blob spheres, I can still spot lots of
> them.
> > I've tinkered with the threshold and strength parameters, but I
> haven't
> > reached any satisfactory values. Any suggestions?
>
> That's much better.  Particle interaction may well be the way to go,
> especially adding viscosity (particles sticking together) as this may
> cure the problem of individual components being visible.  Also it would
> make sure that particles didn't occupy the same space at the same time
> (if that is not the case already).  Good work nonetheless.

Thank you! Yes, currently it is possible for two distinct particles to
occupy the same space. Actually, at the end, when most particles go to the
bottom of the surface, they occupy much less space than it should, because
they're all clustered.

Today in the morning I implemented a routine for particle interaction but
bizarre things began to appear, so I took it out. Maybe today I'll try to do
it again.

Thanks again,

Fernando.

Post a reply to this message

> Today in the morning I implemented a routine for particle interaction but
> bizarre things began to appear, so I took it out. Maybe today I'll try to do
> it again.

What happened that was so bizarre?  Are you using Euler's method for the
simulation (i.e., add a little acceleration to the velocity and add a
little
velocity to the position)?  This may cause the system to gain energy. Is
that
what's happening?

David Buck
Simberon Inc.

Post a reply to this message

> 
> I wrote the code in C, and now it is much faster! I'm beginning to think
> about adding particle interaction. That will slow the algorithm
> considerably, but I think it may be worth the try.

I suppose you know about the megapov plus particle system patch, if not
you might want to have a look.

> I still don't like the appearance of the blobs... it's still too "chunky".
> Even though they are 10,000 blob spheres, I can still spot lots of them.
> I've tinkered with the threshold and strength parameters, but I haven't
> reached any satisfactory values. Any suggestions?

I already suggested to interpolate additional blob components between, but
i'm not sure if that would look good.

Christoph

-- 
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other 
things on: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

"David Buck" <dav### [at] simberoncom> wrote in message
news:3BFDC294.84DA994C@simberon.com...
> > Today in the morning I implemented a routine for particle interaction
but
> > bizarre things began to appear, so I took it out. Maybe today I'll try
to do
> > it again.
>
> What happened that was so bizarre?  Are you using Euler's method for the
> simulation (i.e., add a little acceleration to the velocity and add a
> little
> velocity to the position)?  This may cause the system to gain energy. Is
> that
> what's happening?

Yes, I'm using that method. I didn't know it was Euler's! But the problem
with my interaction routing wasn't really that: I had mistaken the position
vector for the velocity vector!!! :)

I left 3 computers rendering the animation overnight, and it still hasn't
finished! Now I learn that adding reflection to an animation has its
price!!! :)

I hope I can post the result in a few hours.

Fernando.

Post a reply to this message

"Christoph Hormann" <chr### [at] gmxde> wrote in message
news:3BFE452A.712081F4@gmx.de...
>
>

> >
> > I wrote the code in C, and now it is much faster! I'm beginning to think
> > about adding particle interaction. That will slow the algorithm
> > considerably, but I think it may be worth the try.
>
> I suppose you know about the megapov plus particle system patch, if not
> you might want to have a look.

Hi Christoph. Actually no, I've been using POVRay 3.1a and haven't had the
time to experiment with the new beta. Does it have the same feature you're
talking about?

> > I still don't like the appearance of the blobs... it's still too
"chunky".
> > Even though they are 10,000 blob spheres, I can still spot lots of them.
> > I've tinkered with the threshold and strength parameters, but I haven't
> > reached any satisfactory values. Any suggestions?
>
> I already suggested to interpolate additional blob components between, but
> i'm not sure if that would look good.
>
> Christoph

Mmh... don't know... maybe it might help, but I still have to figure out how
to implement it. Now I'm experimenting a bit with of particle interaction.
Actually I'm going to post in a moment my latest experiment. I hope you like
it.

Thanks for the suggestions.

Fernando.

Post a reply to this message

> 
> > I suppose you know about the megapov plus particle system patch, if not
> > you might want to have a look.
> 
> Hi Christoph. Actually no, I've been using POVRay 3.1a and haven't had the
> time to experiment with the new beta. Does it have the same feature you're
> talking about?
> 

No, the particle system patch is not included in 3.5, you can get it at:

http://homepage.mac.com/chrishuff/povpatch.html

Christoph

-- 
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other 
things on: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

news:3bfd37b7@news.povray.org...
> I wrote the code in C, and now it is much faster! I'm beginning to think
> about adding particle interaction. That will slow the algorithm
> considerably, but I think it may be worth the try.
>
> I still don't like the appearance of the blobs... it's still too "chunky".
> Even though they are 10,000 blob spheres, I can still spot lots of them.
> I've tinkered with the threshold and strength parameters, but I haven't
> reached any satisfactory values. Any suggestions?
>
> Well, I hope you like it. And of course, your commentaries are welcomed.
>
> Good luck,
>
> Fernando.
>
Definately shaping up well. Though if you think of water molecules, even a
million blobs would be skimping ;)

Post a reply to this message

"Data" <not@all> wrote in message news:3bfee15a@news.povray.org...
>

> news:3bfd37b7@news.povray.org...
> > I wrote the code in C, and now it is much faster! I'm beginning to think
> > about adding particle interaction. That will slow the algorithm
> > considerably, but I think it may be worth the try.
> >
> > I still don't like the appearance of the blobs... it's still too
"chunky".
> > Even though they are 10,000 blob spheres, I can still spot lots of them.
> > I've tinkered with the threshold and strength parameters, but I haven't
> > reached any satisfactory values. Any suggestions?
> >
> > Well, I hope you like it. And of course, your commentaries are welcomed.
> >
> > Good luck,
> >
> > Fernando.
> >
> Definately shaping up well. Though if you think of water molecules, even a
> million blobs would be skimping ;)

Thanks! I agree with that, but for example, I've seen JRG's animations that
use 10,000 particles (as far as I remember) and it looks much more fluid
than mine, which uses the same number of blob components. But I'm still
experimenting... :)

Fernando.

Post a reply to this message