Scott,
We're significantly off-thread, but I'll answer to the list as long as no
one raises an objection. It IS interesting.
The discussion of whether polarity inversion (flipping the phase), or
whether phase distortion is audible at all has gone on for a long time in
the audio world. The opinions range from (the ear is phase deaf; phase
distortion is not audible" to "phase, and absolute phase are extremely
important and under estimated in terms of their audible significance". The
first comment is attributed to Helmholtz, and the second one is made
frequently in audiophile circles. There is even a book, self-published by
R. C. Johnson, and titled "The Wood Effect; Unaccounted Contributor to Erro=
r
and Confusion in Acoustics and Audio", in which he advocates polarity
inversion as being extremely important. The truth is somewhere in between.
There have been a number of technical papers written about it. Probably th=
e
best one is "On the Audibility of Midrange Phase Distortion in Audio
Systems", which appeared the Journal of the Audio Engineering Society in
September 1982. I have attached the abstract of that paper at the bottom o=
f
this email.
What I have to say is more or less what is expressed in that abstract. I
proved to myself that polarity inversion was audible in 1980 while I was in
the process of doing bench tests on my recently acquired Sony PCM-F1 digita=
l
audio recorder. I observed that there was an audible difference between th=
e
input and output of the device. The first component of that difference was
a simple level difference. When I corrected the level difference there was
still an audible difference, and I eventually figured out that it was
polarity inversion. When I corrected that, I no longer could tell the
difference between the input and the output. That opens another can of
worms.
But as a side issue I decided to conduct a little experiment. I modified a
signal generator to produce audio waveforms which had a significant
asymmetry. I could definitely hear the difference on headphones (remember,
I was working at my test bench). But only when the fundamental frequency
was between about 50 Hz and 400 Hz. Above that frequency the difference
faded. When I tried the same experiment on loudspeakers, in my living room=
,
I couldn't tell any difference between the two polarities. I tried the
headphone experiments on 10 of my coworkers. 9 out of 10 of them scored
100% on identifying the polarity reversal. the 10th one scored only 50%.
It turned out that he was tone deaf!
Which brings me to the quality of the difference. The sound changes timbre
and sometimes loudness. This is not at all what I had expected. I think
that one initial reaction to the concept of polarity inversion is that it
will sound "like I'm standing behind the singer". Or drum, or whatever.
But that's not a really good analogy, because sounds are really different
from behind the source due to the directional characteristics of the source=
.
The reason that polarity inversion is audible is that the ear is NOT phase
deaf, not at low frequencies. I have now left psychoacoustics and am
talking physiology. Recordings from the neurons of the inner hair cells in
the cochlea show that their is output from the hair cells for rarefactions
of the incoming acoustic wave and not for compressions. So the nerve
impulses sent from the ear to the brain actually have information about the
waveform of the acoustic signal. But that only works up to about 800 Hz or
so. For a variety of reasons the phase information stops at about 800 Hz,
and above that frequency the neurons only carry information about the
amplitude of the wave at that frequency.
This goes a little way towards explaining why the signal can have different
timbre or loudness when the signal is inverted. The part of the waveform
that is detected by the ear actually changes when the waveform is inverted.
Of course, for this to happen requires that the waveform HAVE asymmetry. I=
f
it is symmetrical, then turning it upside down results in no change!
Now why only on headphones? It turns out that the presence of reverberatio=
n
in rooms acts as a big phase randomizer. For the most part, when we listen
to loudspeakers in rooms the majority of the sound that we hear is the
reverberation of the room. Only if we sit closer than one meter (in typica=
l
residential rooms) do we hear more speaker sound than room sound. The phas=
e
is so scrambled that signals which were asymmetrical in the recording are
now made to be symmetrical. To demonstrate that that is the natural effect
of randomizing the phase requires a bit more mathematics than I am willing
to go into hear.
So inverting the polarity IS audible, at least on some low-frequency
signals, and when the phase is preserved through the reproduction system.
Most of the time it is NOT audible, because either the asymmetries were
never there in the recording to begin with, or because they were lost in th=
e
loudspeakers or in the room.
Eric Benjamin
On the Audibility of Midrange Phase Distortion in Audio Systems
JAES Volume 30 Number 9 pp. 580-595; September 1982
The current state of our knowledge regarding the audible consequences
of phase nonlinearities in the audio chain is surveyed, a series of
experiments is described which the authors have conducted using a flexible
system of all-pass networks carefully constructed for this purpose, and som=
e
conclusions are drawn regarding the audible effects of midrange phase
distortions. It is known that the inner ear possesses nonlinearity (akin to
an acoustic half-wave rectifier) in its mechanical-to-electrical
transduction, and this would be expected to modify the signal on the
acoustic nerve in a manner which depends upon the acoustic signal waveform,
and so upon the relative phase relationships of the frequency components of
this signal. Some of these effects have been known for over 30 years, and
are quite audible on even very simple signals. Simple experiments are
outlined to enable the readers to demonstrate these effects for themselves.
Having satisfied ourselves that phase distortions can be audible, the types
of phase distortions contributed by the various links in the audio chain ar=
e
surveyed, and it is concluded that only the loudspeaker contributes
significant midrange phase nonlinearities. Confining the investigation to
the audibility of such phase nonlinearities in the midrange, circuitry is
described which enables such effects to be assessed objectivbely fo their
audible consequences. The experiments conducted so far lead to a number of
conclusions. 1) Even quite small midrange phase nonlinearities can be
audible on suitably chosen signals. 2) Audibility is far greater on
headphones than on loudspeakers. 3) Simple acoustic signals generated
anechoically display clear phase audibility on headphones. 4) On normal
music or speech signals phase distortion appears not to be generally
audible, although it was heard with 99% confidence on some recorded vocal
material. It is clear that more work needs to be done to ascertain
acceptable limits for the phase linearity of audio components-limits which
might become more stringent as improved recording/reproduction systems
become available. It is stressed that none of these experiments thus far ha=
s
indicated a present requirement for phase linearity in loudspeakers for the
reproduction of music and speech.
|