Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!uwm.edu!bionet!ames!dftsrv!mimsy!leviathan.cs.umd.edu!ogata
From: ogata@leviathan.cs.umd.edu (Jefferson Ogata)
Newsgroups: comp.music
Subject: Re: What is "QSOUND"
Message-ID: <32445@mimsy.umd.edu>
Date: 4 Apr 91 10:33:19 GMT
References: <91090.130526AARIE@BGUVM.BITNET> <1991Mar31.214440.19139@dgbt.doc.ca> <1991Apr2.114107.1@dev8.mdcbbs.com> <1433@qusunc.queensu.CA>
Sender: news@mimsy.umd.edu
Reply-To: ogata@leviathan.cs.umd.edu (Jefferson Ogata)
Organization: U of Maryland, Dept. of Computer Science, Coll. Pk., MD 20742
Lines: 35

In article <1433@qusunc.queensu.CA> driesse@qucis.queensu.CA (Anthonie Driesse) writes:
|> The way the recordng process is described, it seems that the two playback
|> speakers should ideally be positioned the same way as the microphones
|> used to make the recording.  Then it is also logical why the effect
|> is best with headphones, but I don't see the novelty in the recording
|> process: what makes this different from placing the two microphones
|> in a concerthall to make the recording live rather than from a multi-
|> channel speaker reproduction?
|> 
|> Anton.

I think the point is that there is a speaker for *each* channel that
is being given a discrete location. This refers to channels that have
been recorded by any means whatsoever, not just stereo pairs from two
mikes. This method duplicates not only the relative volume effect
achieved by typical pan controls, but also the time delay between
channels that is rarely if ever present in mixed-down studio
recordings. The human ear actually determines the location of a sound
more by the time delay than by the relative amplitude, believe it or
not. (The delay I'm talking about arises because a sound coming from
your right reaches your right ear first. The ear uses this phase delay
to figure out where the sound is coming from. The sound would also be
*louder* in the right ear, but the delay is more important to the
ear.)

Common mixing boards provide pan controls for adjusting the relative
amplitude, but it is a major pain to try to introduce the delay, and
the actual delay time should vary according to the listener's speaker
placement and location. So as far as I know, no one's ever gone to the
trouble before to put in channel delays for panning effect. This is
effectively what Qsound is doing, if I understand the original poster
correctly.
--
Jefferson Ogata        	        ogata@cs.umd.edu
University Of Maryland          Department of Computer Science