diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Multi-Way (http://www.diyaudio.com/forums/multi-way/)
-   -   Isn't there any 3d audio simulation software out yet?? (http://www.diyaudio.com/forums/multi-way/140049-isnt-there-any-3d-audio-simulation-software-out-yet.html)

JinMTVT 8th March 2009 06:11 PM

Isn't there any 3d audio simulation software out yet??
 
I have searched from google a bit and couldn't find anything of what i was looking for,
as anyone heard of a full 3d wave field simulator ??

i can't see why nobody has come up with that yet,
we've all seen the "tank ripple" java app

wouldn't it be alot easier for us to simulate room modes,
interaction, driver placement etc... with a 3d package ?

could throw in some frequency to see wave propagation and interaction in 3d ???

we could also try complexe arrangement and theories

maybe my head is screwed up also ...or maybe it already exists

i don't think it would be that hard to make such a software, as we can all see how computer 3d rendering
is advanced now with photon mapping and light rays ...

JinMTVT 10th March 2009 12:44 AM

is my question stupid ?
50 views..and no input


i seriously thought i would be placed very quickly
by a member ( or a few ) citing some already existing software solutions ... mmm


i'll then wait humbly for any info ..

Iain McNeill 10th March 2009 12:55 AM

try searching for "finite element modeling"

I think this is still pretty expensive/complex stuff - but then I'm no spring chicken.:)

jason_watkins 10th March 2009 02:27 AM

Light is actually easier. Photon mapping and similar path tracing methods don't have to consider diffraction.

I'm past my depth in math on this topic, but I believe Boundary Element Method is the most applicable, and is still quite expensive computationally. A 3d version of a wave tank would be simpler but even more expensive computationally, since it will be proportional to the volume of space simulated rather than the surface area. I imagine Dr Geddes could also point out acoustic effects that a 3d wave tank would fail to model.

I have been interested in quite some time whether there was a monte carlo simulation method applicable to audio, similar to how participating media is handled in path tracers. My suggestion is that you could use Hyugen's principle similarly to photon tracing, where an initial ray from the source has a probability distribution along it's length of becoming a new virtual point source generating a path in a different direction. Also I think because sound waves interfere significantly at the frequencies of interest, you'd have to accumulate not just energy transfered to the sensor, but also the unwrapped phase/path distance of each sampled path.

Again, being out of my depth mathematically I don't have any clue if this idea can be put on a solid mathematical foundation, and if so what the probability distributions along rays or around the unit sphere for new virtual wavefront sources would be.

soongsc 10th March 2009 02:41 AM

Really depends on how much money you want to spend on buying.

JinMTVT 10th March 2009 03:13 AM

i don't want to spend a dime !!! that's the beauty of internet ... :p

i just want to know if there are any solutions available

i understand that sound wave have multiple properties that are modified along their life and interactions with spatial elements
that makes it expensive for cpu time

but it all depends on "resolution" factor

a software that could resolve basic parameters for a define timeline in a limited space enveloppe
with a basic resolution
could be a crazy tool for us

imagine being able to model your listening room with basic elements ( using even cubes and some basic forms, with material properties )
or beeing able to model a loudspeaker box and place it in an empty room to see diffraction and wave behavior
around/withing the box/room

that would be very very academic

i can play 1-2 hours with the tank ripple app
just trying some patterns and playing with frequency
trying to understand basic stuff
and i believe that beeing able to visualize sound waves
would be so cool, in a wanted situation of course

planet10 10th March 2009 04:51 AM

Quote:

Originally posted by Iain McNeill
try searching for "finite element modeling"
I'm pretty sure there are some open source/free FEMs... learning to use them is a whole nuther kettle of fish.

dave

jlo 10th March 2009 07:48 AM

Quote:

imagine being able to model your listening room with basic elements ( using even cubes and some basic forms, with material properties ) or beeing able to model a loudspeaker box and place it in an empty room to see diffraction and wave behavior around/withing the box/room
I did a kind of auralisation software where you can model your speakers, filters, box, room, and listen to the result moving in the room.
But it is a crude approximation : only paralellepidic room, damping is uniform, etc...
Try it, it's free :
http://www.ohl.to/about-audio/audio-...the-final-cut/
It's missing a manual (but should be easy to use) ;)

MisterTwister 10th March 2009 03:37 PM

I don't see how this 3D wave simulator software would be useful. could work for home theater installers though, to impress customers :)

if you really need to see traveling waves, it can be done in any modern 3D software with particle systems.
"houdini apprentice" is very capable and free. steep lerning curve.

MJK 10th March 2009 03:56 PM

Quote:

Originally posted by JinMTVT
imagine being able to model your listening room with basic elements ( using even cubes and some basic forms, with material properties )
or beeing able to model a loudspeaker box and place it in an empty room to see diffraction and wave behavior
around/withing the box/room

that would be very very academic

That is exactly where I am headed.

First I added the capability to model the driver and port or terminus on a front baffle (or rear for the port or terminus) in discrete locations including the baffle edges and some limited room interaction. That was the lkast set of worksheets I made available.

Now I have upped the solution to account for corner loading, toe-in, and even full rectangular room response. It seems to work reasonably well. These are the worksheest I am using right now.

All of my future speaker designs will be done taking the room into account. You can see the first results, measured and calculated, in my Jordan OB and Goldwood H frame design with the passive crossover that is documented on my site.

I still have a lot of work to do but I think that is the future, the lumped parameter Thiele / Small modeling assumptions take you only so far and this starts to move past those simplifying assumptions.


All times are GMT. The time now is 12:36 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2