In article <mhtr1tcav803aeg1aufat5cmlp2hrrvo4t@4ax.com>, Peter Popov 
<pet### [at] usanet> wrote:

> Does your particle system account for mass?

Yes.


> If so, does it compute the De Broigle wavelength of a particle? Does 
> it then use its duality to compute wave effects?

Um, no...but you can probably do something like that yourself, using the 
age of the particle. I don't know what the "De Broigle wavelength" is, 
so I can't tell for sure...


> Sorry, couldn't resist... but really, it's wave behaviour Greg needs
> to model.

I don't think that is necessary. He said he wanted to use it for things 
like explosions going through hallways, a particle system should be fine 
for that. If he wanted to simulate acoustics, it would be 
different...but that type of explosion is more like a fluid(gas or 
liquid) flowing through a pipe than wave behavior.

It might be possible to do something that goes the other way, sort of a 
cross between radiosity and the proximity pattern. You would specify a 
target object and a recursion level, and it would send out samples to 
determine the visibility of the object from any point, even the effect 
around corners. However, this would be incredibly slow...though a sort 
of cache could be used for reusing data, like radiosity has, I don't 
know how to implement that. And the particle system approach could 
probably come closer to simulating wave behavior.

Hmm, maybe some sort of 3D grid, similar to a voxel image, could be set 
up around the immediate area containing all the "free" space, this could 
then be used in waveform simulations...it could be faster, because no 
recursive tracing would need to be done, but would eat memory like crazy 
in large spaces with high resolution. An octree-like structure could be 
optimized for large areas, but areas that are mostly "free" would 
probably use memory best when a simple array is used.

-- 
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/

<><

Post a reply to this message