Xref: utzoo sci.electronics:2411 rec.music.synth:2689
Path: utzoo!utgpu!water!watmath!onfcanim!dave
From: dave@onfcanim.UUCP (Dave Martindale)
Newsgroups: sci.electronics,rec.music.synth
Subject: Re: guitar audio signals
Message-ID: <15567@onfcanim.UUCP>
Date: 4 Mar 88 05:12:32 GMT
References: <502@m10ux.UUCP> <22670@yale-celray.yale.UUCP> <8948@ism780c.UUCP> <1165@mtunb.ATT.COM> <306@nvuxj.UUCP>
Reply-To: dave@onfcanim.UUCP (Dave Martindale)
Organization: National Film Board / Office national du film, Montreal
Lines: 25

In article <306@nvuxj.UUCP> abeles@nvuxj.UUCP (J. H. Abeles) writes:

>The RIAA curve is chosen
>to approximate the opposite (i.e., a rolloff proportional to freqency).
>Its several 3 dB break points are chosen to approximate this 1 over
>omega rolloff.  A single pole filter is 1 over omega in voltage but
>I think not in power, or something like that.

Magnetic phono cartridges basically respond to velocity.  If you cut
records so that the cartridge output was flat, the *amplitude* of the
groove modulations would drop by a factor of 2 per octave.  The low
frequencies would be untrackable and the highs would be buried in noise.

If you apply a 6 dB/octave boost in recording and the reverse in playback,
you are now producing a groove whose amplitude is constant with frequency.
This is much better, and also works nicely with ceramic cartridges, which
are amplitude-responding.

Unfortunately, the 10-octave range of input signal requires a 60 dB
boost/cut of the the high frequencies with respect to the low, and you
want a 60 dB dynamic range for the music on top of that - that's asking
a lot of the electronics.  To reduce this problem, the RIAA characteristic
reverts from constant-amplitude to constant-velocity for about 2 octaves
in the middle of the range (0.5 - 2 KHz?).  This produces the "shelf"
in the equalization curve.