Adapting lux to render sound?

[thomas]

Hi,

Do you think it would be possible to adapt lux to render other phenomena, in particular sound? So instead of sampling light sources with light spectra, sample sound sources with frequency spectra and trace them back to the ear, thus spitting out a sound file. Due to differences in scale and speed of the waveforms certain effects may play a larger role, such as diffraction, phase and time. But do you think something like this would make any sense at all? Does anyone know of some examples that actually do this? It would help me out a lot if anyone of you could shed their light (or sound) on this subject.

Take care,
Thomas

[SATtva]

I think longer wavelength are more prone to wave interference. Sound doesn't exhibit wave-particle dualism (which makes it possible to trace photons like simple rays), so you'll have to base the simulation on waves. So no, I don't think this is possible with Lux without extensive modifications.

[tomb]

SATtva is right - for longer wavelengths diffraction is very much apparent and FEM methods are usually used to solve the problem. For mid- to high-frequency sound, I've read about something called "phonon tracing", look it up in google It should be possible to extend PBRT or Lux to do such phonon-mapping, but I kind of doubt you'll find anyone with time/inspiration to do it (within the lux team at least)...

[Lord Crc]

shameless plug: if you decide to go the fem route, be sure to check out FEniCS, http://fenics.org, a very powerful and fast GPL framework for solving partial differential equations (using FEM).

[thomas]

Thanks for the replies all even if it are the replies I sort of feared.

I am interested in the psychoacoustic phenomena in architecture and a way to evaluate them, particularly in relation to human movement, so realism from a scientific point of view is not the very most important, an indication would already be nice. My ultimate goal is to render auditory walkthroughs, so It seems to me that FE methods with their limited resolution would pose some problems.

But thanks again, I´ll look into it some more and might as well end up with FEM and FEniCS after all. Btw, tomb, I did not mean to imply that I wanted anyone of you to do this, sorry if my question was posed in that way.

[tomb]

Thanks for the replies all even if it are the replies I sort of feared.

I am interested in the psychoacoustic phenomena in architecture and a way to evaluate them, particularly in relation to human movement, so realism from a scientific point of view is not the very most important, an indication would already be nice. My ultimate goal is to render auditory walkthroughs, so It seems to me that FE methods with their limited resolution would pose some problems.

But thanks again, I´ll look into it some more and might as well end up with FEM and FEniCS after all. Btw, tomb, I did not mean to imply that I wanted anyone of you to do this, sorry if my question was posed in that way.

Not at all! I just said it so you didn't get your hopes up too much If this is something you really need/want, I think I'd go to a university and propose it as a semester assignment or perhaps a master's project or something like that.

[thomas]

haha ok great, well regarding the university thing, I am actually in the middle of investigating my graduation project (architecture) at the moment, hence these questions

[tomb]

Hehe, sounds pretty cool! I'm sure we'll try to answer any questions you might have in the core engine forum if you do try to implement such an integrator. Out of curiosity; is it the visualisation or the "auralisation" that's your prime goal?

[thomas]

Thanks tomb, I appreciate that, let me answer your last question with an excerpt from an introduction to my project:

The aural qualities of architecture are often neglected. This is partly due the fact that these qualities are hard to evaluate, hard to communicate and, most of all, hard to remember. And even when aural qualities are evaluated, often only at a functional level of e.g. reverberation times, the auditive qualities that define a sense of space, a sense of orientation, a sense of a spatial identity are left out. Tools to evaluate the aural qualities of architectural space, especially in relation to movement of the beholder and sounds from its surroundings, are not generally available.

The aim of this project is, on the one hand, to create a working prototype of such a tool. On the other hand, to create an architectural work in which her aural qualities explicitly play a major role.

[thomas]

Well, I did it! I wrote my own ray-tracer to simulate the propagation of sound. As already noted by all of you, it is not physically correct, but I think it can give a nice impression of the auditory experience of a space. My primary goal was to give architects a tool to investigate the aural experience that unfolds as visitors navigate through the building. But I think it has evolved into a tool that it useful to everyone with an interest in computer graphics, film making and special effects.

In the end I did not use the luxrender sources, mainly because I am not smart enough to understand the code, but did find inspiration in the way lux is integrated into Blender and its GPLness...

I invite everyone give it a try, it is called E.A.R: Evaluation of Acoustics using Ray-tracing. I tried to make things user-friendly and write documentation, but man that's a lot of work. If you have any suggestion or questions you know where to find me. Get your copy of EAR from one of the two websites below:

The code is hosted at github: https://github.com/aothms/ear
And there is a website called http://www.explauralisation.org/

I have created two example videos, material can be downloaded from the websites. But I'm sure you can create way better work with EAR.
http://youtu.be/b-HlI-zbqzU
http://youtu.be/YnzWnTX8eZE

Kind regards,
Thomas