Tim Nikias v2.0 wrote:
> If you could explain a little more in detail
> what it is supposed to do, that would be helpful.
> Is it some algorithm with which I could calculate
> the movement of smoke or fire (due to the way hot
> particles move), or am I on the wrong track?
The algorithm itself is very general. You may have heard of the Verlet 
Algorithm from which the leapfrog algorithm is a variant. Basically it's 
a numerical integration method like the Euler Algorithm, but it has some 
advantages over the Euler algorithm. The energy conservation is better 
and the algorithm runs stable in the other direction.
Such programs can be used for very complex simulations for molecular 
dynamics. There is a great interest for such simulations in the 
industrie, especially the chemical industrie. With enough effort you 
should be able to model smoke or fire physical accurate, but it may not 
be suited for animation purposes because of speed. Actually I think a 
"fake" could be suited much better.
My animation can't show really, what the program is capable of, but I 
thought it would be nice for a presentation. Note, that the program 
calculates the distances of the particles with periodical walls, but the 
particles itself move freely in the room. There is a cut radius, so that 
forces from far particles are not calculated.
You may have a look at the group page for more general information on 
molecular dynamics (german):
http://constanze.materials.uni-wuppertal.de

> 
> Whats the parsing time (render time shouldn't
> be that high with this setup, obviously :-)?
> 
I can't remember the parsing time, but the whole time for one picture 
(1280x1024) was 27s. I thought about using focal blur, but didn't had 
the patience ;-).

Florian

Post a reply to this message