Andy wrote:
> > I'm talking about actual waveform analysis compared to the origional
> > sound.

Tom replied:
> Ah, so who cares what it sounds like - we'll tell you what you should be
> thinking.

Dan responded:
I'm not getting what you are arguing.  Are you saying that fidelity to the
sound put out by the folks who made the CD is not an important part of a
player?

Fidelity is not always fidelity. There a lot of psychoacoustic tricks that can be
used to create compression that loses a lot of data and therefore looks quite
a bit different from the original sound in a waveform analysis, but does not
affect the perceived sound (or doesn't affect it much). In fact, compression
schemes such as MP3 would be impossible without using some of those
psychoacoustic phenomena. The trick is simply to get rid of sound that
most people don't perceive (or perceive only weakly) anyway.


For example, parts of the sound that would cancel in the air due to phase
cancellation can be reduced or removed in some circumstances, as can
sound that is in one channel but is masked by sound from the other
channel.  Codecs that do this kind of compression typically expect that
you will be listening through speakers, as this kind of cancellation would
not occur if you were wearing headphones (and since most professional
audio mixes are done using speakers as opposed to headphones, this
kind of lossy compression would actually give headphone listeners a
more accurate version of the recording than if they listened to the
original uncompressed version through headphones).

Wikipedia also has an entry that offers some additional description of
this with links.
http://en.wikipedia.org/wiki/Psychoacoustic_model

I haven't spent a lot of time listening to different compression schemes
to see which one I think sounds most like the original, but I would
guess that to some extent it would vary from person to person.  Each
person perceives sound a little differently, hearing some frequencies
more strongly than others, and I would guess that some compression
schemes will remove sound that some people perceive strongly
(resulting in them thinking that particular compression scheme sucks)
whereas other poeple may not perceive the frequencies being removed
as much and so therefore think the compressed audio sounds much like
the original, uncompressed version.

I would also guess that much sound removed by various compression
schemes (particularly MP3) is very high frequency sound, which many
people can't hear anyway.  High frequency hearing loss is the most
common kind of hearing loss, and can affect perception of music
long before it becomes an issue in being able to understand speech.

So I guess the simple answer to your question, Dan, would be that
fidelity to the sound of the original does not always equal having the
waveforms look the same or similar, and the compression scheme
that looks most like the original will not always be the one that
best preserves the sonic impression of the original.  When it comes
to sound, I trust my ears much more than my eyes.

Reggie Bautista

_________________________________________________________________
See when your friends are online with MSN Messenger 6.0. Download it now FREE! http://msnmessenger-download.com


_______________________________________________
http://www.mccmedia.com/mailman/listinfo/brin-l

Reply via email to