Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site eosp1.UUCP
Path: utzoo!linus!philabs!cmcl2!floyd!harpo!ulysses!allegra!princeton!eosp1!lincoln
From: lincoln@eosp1.UUCP (Dick Lincoln)
Newsgroups: net.audio
Subject: Re: RE: Extra speakers
Message-ID: <789@eosp1.UUCP>
Date: Wed, 11-Apr-84 10:20:53 EST
Article-I.D.: eosp1.789
Posted: Wed Apr 11 10:20:53 1984
Date-Received: Fri, 13-Apr-84 01:23:56 EST
References: <2688@rabbit.UUCP> <36@brl-vgr.ARPA>
Organization: Exxon Office Systems, Princeton, NJ
Lines: 34

From rabbit!jj

>> Ok, Ron, let's put it this way:

>> How do you get the second speaker to vibrate at the different
>> frequency?  How do you manage to NOT have the beats between the two
>> frequencies, ONLY a pitch shift?  How do you propose to make this
>> system time-varying?

From Ron Natalie

> Personally, I think that it's just as likely as doppler shift or any
> other effect you are going to get from a 3" speaker.

Sorry, Ron, but linear system theory is on jj's side on this one, and I
believe we may assume that linear system theory is applicable at the
minute power levels possible for "passive" acoustical energy storage
into a 2-4" speaker connected to "dead" electronics.

This theory states in effect that it is impossible to produce a
frequency pitch shift *in a relatively stationary object* (eliminating
Doppler effect) without a significant non-linearity, such as the "super
het" signal multiplier designed in receiver front ends specifically to
produce just this frequency shift translation through "beat"
frequencies from the incoming antenna signal and a single oscillator
tone.  Note, however, that a time *delay* (increasing phase shift with
increasing frequency) is quite possible with just "linear" components -
happens all the time.  This does *not* produce a pitch change, though.

There is no reason to accept a pitch change without a credible,
significantly non-linear, "passive resonator" model, or some convincing
psychological data indicating humans interpret a time delayed and
frequency filtered echo combined with its original signal as a pitch
change.