NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/rec/audio/17589

Internet Message Format | 1992-12-31 | 3.5 KB

Path: sparky!uunet!noc.near.net!news.bbn.com!NewsWatcher!user
From: [e-mail] (Kerry Shetline)
Newsgroups: rec.audio
Subject: Re: polarity inversion
Followup-To: rec.audio
Date: 31 Dec 1992 06:58:06 GMT
Organization: BBN
Lines: 61
Distribution: world
Message-ID: <shetline-311292015736@128.89.19.85>
References: <[e-mail]> <[e-mail]>
NNTP-Posting-Host: bbn.com

In article <[e-mail]>, [e-mail]
writes...
>
>   Just how does polarity inversion affect the sound of a stereo?  I have
>never had a system that had this feature.  I have no idea what it does.
>Could somebody explain?  thanks!

Sound waves in air consist of radiating patterns of compression and
rarefaction (increases and decreases in pressure). Suppose a microphone
picks up a compression while making a recording. If absolute phase is
maintained throughout the audio recording/playback chain, a loudspeaker
reproducing this sound will create a corresponding compression at the
listener's ear. If polarity is reversed, the microphone's compression will
result in a rarefaction at the listener's ear.

Well, so what? If one channel in a stereo pair is in phase and the other is
not, it makes a big difference. The two speakers fight each other and end
up canceling out much of the sound. But if both match...

Then this is a matter of religious war. Can the human ear detect absolute
phase? Most tests of human subjects, as well as the design of the human
ear, would tend to rule this out. The ear acts as a spectrum analyzer, and
as far as spectral analysis goes, absolute phase does not matter. The
measured amplitude of any given frequency band can not be affected by
polarity.

However, under controlled circumstances, I believe a few rare individuals
have been shown to do better-than-chance at recognizing polarity inversions
in signals that they have carefully listened to many times. I don't
believe, however, that there was any consistent preference, or any ability
to tell which polarity was "correct" in these tests. The differences might
be explainable in terms of the equipment used in the tests. A particular
speaker might, for instance, not create exactly the same magnitude of
pressure differential when fed +X volts or -X volts. This could change
sounds in a subtle way that has nothing to do with actually hearing
inversions of polarity.

With the way many recordings are made, only minimally-miked classical
recordings are likely to even have recoverable absolute phase anyway. With
something that has been multi-miked, multi-tracked, mixed, edited, and
recorded in separate sessions with different setups, absolute phase is
completely jumbled and mixed with inverted phase -- different for different
voices and instruments, on different tracks or parts within tracks.

So, except for the stereo channel mismatch mentioned above, absolute phase
comes down to (IMHO):

1) YOU probably can't hear it.
2) Maybe no one can hear it.
3) If you can hear it, you probably won't care.
4) If you can hear it, and you do care, you often can't get it.
5) If you can hear it, you do care, and you can get it, you may not know
when you've got it. (Do I push the little button, or don't I?)

There are, of course, people who make extravagant claims as to the enormous
effect of polarity. Some person on this news group recently spoke of CDs
that went from unlistenable to wondrous with the touch of a polarity
switch. The answer to this phenomenon is not likely to be found in audio
literature, but somewhere is DSM III...

-Kerry