Reply-To: "Kevin Aylward"
From: "Kevin Aylward"
Subject: Re: Bullshit wins v. science
X-Newsreader: Microsoft Outlook Express 6.00.2800.1106
X-Inktomi-Trace: public1-pete2-5-cust19.pete.broadband.ntl.com 1036695558 24825 22.214.171.124 (7 Nov 2002 18:59:18 GMT)
Date: Thu, 7 Nov 2002 18:59:13 -0000
NNTP-Posting-Date: Thu, 07 Nov 2002 18:59:18 GMT
"Mark Zenier" wrote in message
> In article ,
> Kevin Aylward wrote:
> >"Mark Zenier" wrote in message
> >> In article ,
> >> Kevin Aylward wrote:
> >> >This is a misconception. The ear is *absolutly* *insensitive* to
> >> >A static phase shift is quite undetectable. It is a Fourier
> >> Er, are you deaf in one ear? Both ears combine to be sensitive to
> >> phase in some of the frequency range.
> >> Take a mono signal and feed it into headphones and you'll get a
> >> sound source that's small point in the center of your head. Swap
> >> wires on one ear's speaker, and the perception changes to a diffuse
> >> sources outside your head, on either side.
> >Ahmmm. Again, This is a misunderstanding. As I *keep* pointing out,
> >is possible to detect phase difference by *comparison* of one signal
> >with another. I have made this quite clear that adding/subtracting
> >signals will result in frequency response effects. It is the
> >spectrum effect you detect, not the phase.
> I'm having a semantic quibble about the dogmatic statement you made.
There is nothing dogmatic about it at all. Its a recognition of a basic
principle of physics that is simply not defeatable.
To detect phase difference you *must* have a reference. Phase on its own
is quite *meaningless*, so it quite impossible, in principle to detect
absolute phase. Explain how you can measure 5V. What does 5V mean?
> The point is that "the ear" has two senses, one that percieves WHAT
> are hearing, the other WHERE the sound is coming from. I contend that
> the latter uses phase, or time of arrival in the millisecond time
Please explain a physical mechanism that allows a measurement of
The point is, is that you are not paying attebtion to what I have wrote.
> (I don't remember a reference to humans, but I remember science media
> coverage of some little pencilneck bird down in Australia (a plover
> or something) that some researchers were trying to figure out how it
> could get direction information from a wavelength ten times the size
> its head).
> >However, this is not relevant at all to the statement that the ear is
> >quite insensitive to static phase shifts. That is, take a signal.
> >it through a phase shifter, either constant or phase varying with
> >frequency. Listen to the output *on its own*. You will simply not be
> >able to hear any difference.
> Try this. Get a monophonic noise source.
Ho.. Humm... Start of irrelevance again...?
> White noise, or better
> something that has a nice crackle or rustle. (Something from our
> evolutionary past indicating dinner). Add in a nice 400 Hz, or
> thereabouts sine wave, to add a bit of depth. Feed this into
> one ear. For the other ear,
>feed the signal though a allpass filter
> or time delay, that can be switched into the circuit or bypassed.
> Put on the headphones and switch the filter in and out.
Jese wept dude, your mixing two signals again, all bets are off.
> WHAT you're hearing may not change. But WHERE the sound seems to
> come from sure will.
Nope. If you have a simple time delay the same in both earphones it will
make no difference at all. You cannot tell if someone is speaking to you
from the US or the UK, excepting poor echo cancellations and other
blemishes. Or more obviously, listening to Neil Armstrong on the moon
sounds no different then if he was in the next room on the phone.
> Or just try to listen to Midnight Oils "Diesel and Dust" CD on
> headphones. Enough platform motion to make you seasick.
I am not familiar with this CD, so I can't really comment until I know
what is being done. Invariable, I would guess that it will be mixing two
signals. Mixed signals automatically result in *frequency* response
effects. e.g. a mixed delay will produce a comb shaped frequency filter.
I have stated all the time that you can detect phase *if* you are
comparing it with a *reference* phase. This is obvious and trivial.
However, this still has no relevance to that fact that the ear itself is
insensitive to absolute phase. As I said, a wideband music signal feed
through an allpass filter, sounds exactly the same as without the
filter. Try it.
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.