Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/3/84; site teddy.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!gamma!epsilon!zeta!sabre!petrus!bellcore!decvax!genrad!panda!teddy!rdp
From: rdp@teddy.UUCP
Newsgroups: net.audio
Subject: Re: CD player differences
Message-ID: <1177@teddy.UUCP>
Date: Wed, 21-Aug-85 17:03:15 EDT
Article-I.D.: teddy.1177
Posted: Wed Aug 21 17:03:15 1985
Date-Received: Sat, 24-Aug-85 18:27:52 EDT
References: <456@olivee.UUCP> <4150@alice.UUCP> <463@olivee.UUCP> <1613@druil.UUCP>
Reply-To: rdp@teddy.UUCP (Richard D. Pierce)
Organization: GenRad, Inc., Concord, Mass.
Lines: 61
Summary: 

In article <1613@druil.UUCP> lat@druil.UUCP (TepperL) writes:
>> ... I still feel that
>> differences in output level would only have seriously affected the
>> first few minutes of hearing either player.  The other differences,
>> which I described at some length in the original article, were of a
>> nature, and heard over sufficiently long periods of time, that I can't
>> attribute to level mismatches.
>> 
>> 	- Greg Paley
>
>Since you're so sure that matching levels doesn't matter, next
>time you compare CD players, why don't you match levels (since it
>won't affect the outcome) anyway and be done with it?
>
>Certainly as some point it will make a difference.

You betcha it will make a difference!

I have been watching this discussion about CD's with some amusement
for some months now, and the latest argument about level matching
I found particularily amusing because of my own experiences when I
was in the retail business.

That is, unitl I REALLY remembered what went on...

I was involved with a store (no defunct) which did its utmost to provide
fair, accurate, well-informed information on not only the equipment that
we sold, but ALL equipment. We attempted to be as honest as possible. 

However, it seems that quite a few stores in the Boston area were either
unaware of the fact that even minute differences in level could be mis-
interpreted as qualitative differences, but there were some who quite
conciously exploited that fact to drive a sale home. In fact, one fledgling
chain, who wanted us dead, adopted this as a stated, but hushed, matter of
company wide policy, "If the customer doesn't like one of our brands, then
turn it up until he does". It was very difficult to survive in the face of
this sort of unscrupulous behavior, especially when the competition
accused US of this activity!

To prove our point, we performed a little experiment on some of our
regular visitors. We ostensibly set up a comparison between a new Danish
loudspeaker and Large Advents (the then rage in the Boston area), and had
people compare them via A-B switching. In most cases, the listeners who
liked the Advents picked them out with little hesitation, saying how much
better they obviously sounded. 

In fact, what we had set up was a comparison between the Danish speaker,
and the Danish speaker .5 db louder! Even though the two were absoloutely
identical, except for output level, people attributed the difference to
large quantitative differences between two different loudspeakers. Note that
of the 10% or so that did not pick the Advent, they were undecided. I don't
recall ANYONE saying "Hey! that's the same loudspeaker, only louder!"

So, Mr. Paley, pontificate not on the unimportance of reducing variances
in tests, it is real damned important! Unless you can point to specific
experiments that support your claim of either the non-audibility, or the
lack of long-term importance of mis-matched levels, then I would, too,
recommend, especially in the light of real experiments like the one I
described, make damn sure your levels are as closely matched as possible.

Dick Pierce