NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/rec/audio/19178

Text File | 1993-01-23 | 3.7 KB | 71 lines

Newsgroups: rec.audio
Path: sparky!uunet!elroy.jpl.nasa.gov!swrinde!emory!sol.ctr.columbia.edu!eff!world!DPierce
From: [e-mail] (Richard D Pierce)
Subject: Re: a point about the Bose thread
Message-ID: <[e-mail]>
Organization: The World Public Access UNIX, Brookline, MA
References: <[e-mail]>
Distribution: na
Date: Fri, 22 Jan 1993 21:49:14 GMT
Lines: 59

In article <[e-mail]> [e-mail] (Todd Premo) writes:
>I have a simple question to those who are knowledgable about this stuff:
>
>    The Bose 901's are attempting to do something that is
>already being taken care of with the recording- but I have a problem
>with this.
>
>    Wouldn't it be almost better to make the recording in a studio
>that did not allow (or at least minimized) the reflections reaching the
>mike and then reproduced them on a speaker that took care of sending the
>sound in similar directions as the original source?  Of cource this
>would be a different set of reflections in your home than it would be
>in the auditorium, but it would be equivilnt to having that same group in
>playing in your living room.  natural.
>

No, and this was one of the points I brought up before. In a real
live listening situation, typically you are immersed in a complex mix of
direct, reflected (usually just a reflection or two), both of which belong
to the early arrival stuff, and reverberent sound. Bose oft quotes the
number of 11 percent direct, 89% REVERBERENT sound. Then he talks about
the 901 generating a sound field in your living room that's 11 percent
direct and 89% REFLECTED.

There are two flaws in his magic leap of logic. First, anyone of a number
of studies could come up with numbers more or less than Bose's. But far
more serious a flaw is his equating REVERBERENT with REFLECTED. The two
are completely, utterly, totally different. The reflected stuff out of
901's usually arives after 1 or 2 reflections with delay times typically
less than 10 milleseconds and generally from the direction of the speakers
with amplitudes and delays such that they effectively scramble whatever
localization clues exist in the music, whereas the reverberent field in a
live situation is arriving with innumerable generally large, random delays
and generally exponentially decaying amplitudes and contributes little to
the localization cues or scrambling thereof.

Further, there is another aspect about the 901's I've not brought up. I'll
illustrate with an example I personally measured. 901's in a new room, set
up precisely according to the Bose religion had a tremendous suckout at
about 150 Hz (like 30 dB deep). The wall that they were against was shared
by a bedroom. The owners of the house decided to add sound insulation in
that wall by drilling holes and having cellulose insulation blown in. The
transmission through the walls overall did not improve much, but the
suckout in the listening room went away.

Why? Because in the recommended setup, the rear facing drivers couple very
stringly (almost mechanically) to the wall behind them. IN this case, the
wall was constructed out of normal drywall on 16" OC 2x4 studs. Other
walls in the house exhibited a very high-Q drum-like resonance at 150 Hz,
which was simply the panel resonance of the drywall hanging on the studs.
Other speakers that were not so tightly coupled had nowhere near the problem.

While filling the walls did make the problem less, it did not last.
Eventually the cellulose settled and blew the drywall off the nails!
--
|                Dick Pierce                |
|    Loudspeaker and Software Consulting    |
| 17 Sartelle Street   Pepperell, MA  01463 |
|       [phone] (Voice and FAX)      |