[linux-audio-user] Audio 3-D Demo --- Any Interest in Software?

davidrclark at earthlink.net davidrclark at earthlink.net
Wed Jan 7 11:03:15 EST 2004


> I guess this is where this whole discussion so far has left me confused.
> There certainly could be 2, or 3, or more CDs (or LPs) released. The
> question is how does a consumer know which one to purchase? Does he/she
> purchase only for headphones? Do they purchase for bad living rooms? Great
> tuned stereos? Cars?
> I don't see how this can work practically. 


I agree completely than eight different versions of a CD wouldn't work.
But this isn't purely a discussion of headphones versus speakers.  Yes,
there appears to be some interest in 3-D audio, but it isn't limited to
headphones.  My demo is for headphones, yes, but the same thing can
be done with speakers by taking another step with the wave equation,
and that is to do retrodiction in the listening environment (or in a
generic one).  

Many people already balance all of the speaker environments you mention;
they just don't include headphones.  So I don't view inclusion of headphones
as another environment to become sensitive about to be a big deal.  It
would seem to me that as soon as people become accustomed to better
headphone listening, they may prefer it just like I do, even though I
used not to.  As headphones become more prevalent --- as they almost 
certainly have lately --- mixes that favor headphones or at least include
them may be more popular than those that don't.  (I just saw Joe's comment
which tends to support what I'm saying here, at least as I read it.)

But again, what I've done is by no means limited to headphones, despite
my demo.  The headphone solution is a reference point where everyone
will hear pretty much the same thing, assuming they have decent headphones.

In addition, what I've done also includes more realistic instrument
generation, and to some degree, integration of instruments and environments.
It's a more physical way of thinking about all of it as opposed to the
DSP-oriented approach (of the previous century, now long past...).


Maybe that's the key point I should emphasize: I'm developing more
physical ways to model audio, including effects of rooms (including
fictitious rooms, by the way), instruments, impulse generation (modelling
actual playing such as picking a string on a guitar), etc.  I believe that
this physical modelling may very well make creating and mixing music 
easier and more realistic sounding.  My demo is a small part of that to
show that 3-dimensional rooms can be modelled to obtain headphone 
solutions which sound better than the usual mix --- at least they do 
to me.  This headphone demo led me to wondering if people would begin to
prefer headphones over speakers, like I now do.  Previously I didn't.


By the way, I'm still curious about the questions I asked about those
IR's you're working with.  Or are you proscribed from discussing them?
I'm sure others would also like to hear about what you're seeing if you
can discuss it.  Thanks for any discussion about them!


More information about the Linux-audio-user mailing list