[ardour-users] Audio engine: reality or myth ?/ what about Ardour ?

Steve Wahl steve at pro-ns.net
Wed Mar 29 08:13:05 PST 2006


On Tue, Mar 28, 2006 at 10:35:45AM +0200, Jazzride2006 wrote:
> Alex Megalokonomos a écrit :
> > I think what he is asking is whether there is any truth to the concept of an audio engien aloen altering the sound/quality
> 
> exactly :)
> 
> just playback..., will there be a difference in sound/quality? Yes ...I 
> think there is....Tiny and purely subjective mind you though...
> 
> 
> So ? :)

For playback, if two different programs feed the same data to the
sound driver, and both do it with no underruns, no starvation of the
data, the only source of a difference in the resulting sound should be
a difference in the electrical noise that the computer generates while
running the particular program -- and only to the extent that your
(or my) poor hardware doesn't keep the noise out, right?

Howwever, if the programs are doing ANY math on the source information
(mixing wav files, even changing volume, sample rate conversion,
etc.), I'm quite certain there's room for subtle errors, or (lack of)
"black magic" techniques such as dithering, that can make them sound
subtly different.  (If you mix two 16 bit wave files to a 16 bit
output files, once in a program that does 16 bit integer math, and
once in another program that does 32-bit floating point, it's likely
that you won't get exactly the same results.)

So I believe it is possible for one "audio engine" to be of better
quality than another, or at least different sounding.

Is that correct, and does it answer the question?

-- 
Steve Wahl    steve at pro-ns.net

"If you haven't tried NCSA Mosaic to travel the Internet, then you are
missing the best way to experience the Internet...Its so good, I think
we should make a WWW server [here], and get a [256kbps] connection to
the Internet."  -- Rick Richardson, 9 Aug 1993



More information about the Ardour-Users mailing list