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!linus!philabs!cmcl2!seismo!harvard!talcott!panda!teddy!rdp
From: rdp@teddy.UUCP
Newsgroups: net.audio
Subject: Re: equalizers
Message-ID: <986@teddy.UUCP>
Date: Tue, 23-Jul-85 10:33:54 EDT
Article-I.D.: teddy.986
Posted: Tue Jul 23 10:33:54 1985
Date-Received: Thu, 25-Jul-85 04:53:11 EDT
References: <7351@watdaisy.UUCP> <1545@watdcsu.UUCP>
Reply-To: rdp@teddy.UUCP (Richard D. Pierce)
Distribution: net
Organization: GenRad, Inc., Concord, Mass.
Lines: 42

In article <1545@watdcsu.UUCP> herbie@watdcsu.UUCP (Herb Chong [DCS]) writes:
>In article <7351@watdaisy.UUCP> gvcormack@watdaisy.UUCP (Gordon V. Cormack) writes:
>>It is my impression that almost all differences between reasonable
>>quality stereo components can be attributed to differences in
>>frequency response across the audio spectrum.  Therefore, I have 
>>decided that I should buy a multi-band equalizer.
>
>a controversial statement, but one of which i happen to believe
>to be true.
>

Yes indeed, a controversial statement, but it does not have a whole bunch
going for it. Much of the coloration that we hear can be attributed to
frequency response anomolies, but simple linear equalization is most often
completely inadequate to the task.

Take the problem of room resonanaces, something which people usually by
equalizers to correct. You can measure errors in the steady-state frequency
response and then equalize for these errors, and the steady state response
might look better. But, what happens in such resonances (and those associated
with speaker problems as well, such as cabinet resonances, cone resonances,
etc.) usually result from long term energy storage and subsequent release
at different ffrequencies. For example, there is the tale about the early
days of computerized speaker analysis where there was a certain speaker
which measured fine under steady state conditions, but people complained
of it being very "brittle" or "sizzly". It was later discovered that after
the driving signal had ceased, the speaker's output decayed normally until
about 2 or 3 milliseconds after the cutoff, when it started singing away
at 7 Khz for some 20 milliseconds!. An equalizer cannot in any way cope
with this sort of anomoly. 

WHat equalizers are best for are taking care of mis-equalized records, 
poorly designed crossover networks, slight errors in RIAA phono equalization
and so forth. WIth the exception of the first, these items aren't the province
of an equalizer to correct, they should have been done correctly to begin
with.

In short, my advice to the vast majority of equalizer customers is: save
your money, and use it to upgrade other components that are at fault. 
Solving one frequency response error with another is non-productive.

Dick Pierce