source file: mills2.txt
Date: Fri, 16 May 1997 13:12:59 +0200

Subject: JI Composition

From: Mckyyy@aol.com

Someone asked if there is a general method of composition in JI. 
I'm going to be lazy this morning and not look up the specific
quote.

Considering that JI composition has been going on for at least
centuries, probably millennia, and if you consider birdsongs and
such to be compositions, perhaps aeons, I would imagine there are
several established methods.

I have used a method based on least common multiples.  Broadly
speaking, the smaller the least common multiple of a chord, the
more harmonious it sounds.  The LCM of a diad with the ratio of
3:2 is 6.  The LCM of major/minor triads is 60, etc.

My theory was that chords with smaller LCM's create patterns that
repeat in a shorter time.  Paul E was kind enough to point out
that theory was not exactly accurate, in the way I was using it,
nevertheless, it seems to be of some use.

So now there is the 'pure LCM' theory and the 'modified LCM'
theory.  I wrote a music sequencer, FasTrak, which uses the 'pure
LCM' theory.  I hope to write another someday that uses both.

I believe FasTrak can still be downloaded by FTP from mmcky on
AOL.  It requires an IBM pc and a basic sound blaster.  It needs
to be run in DOS.

FasTrak comes with a bunch of utilities for analyzing JI scales
and creating chord sets that are useful for JI compositions. 
These can be useful to someone studying JI no matter what the
composition environment.

On the subject of sampling and JI.

As sampling is usually implemented, the tuning accuracy of an
instrument may vary according to the particular sound being
reproduced.  That is because samples of different lengths are
used for different instruments.  In the world of
'transcendental/irrational' tunings, where approximations are all
that are available, this makes little difference, but for exact
JI, it is a severe drawback, IMHO.

The lack of tuning accuracy among current instruments, and the
lack of disclosure by synth manufacturer's of tuning accuracy
specs, makes the whole tuning discussion much more difficult than
it needs to be.

Marion

Received: from ns.ezh.nl [137.174.112.59] by vbv40.ezh.nl
           with SMTP-OpenVMS via TCP/IP; Fri, 16 May 1997 14:08 +0200
Received: by ns.ezh.nl; (5.65v3.2/1.3/10May95) id AA03692; Fri, 16 May 1997 14:08:13 +0200
Date: Fri, 16 May 1997 14:08:13 +0200
Received: from ella.mills.edu by ns (smtpxd); id XA03690
Received: (qmail 12467 invoked from network); 16 May 1997 12:08:09 -0000
Received: from localhost (HELO ella.mills.edu) (127.0.0.1)
  by localhost with SMTP; 16 May 1997 12:08:09 -0000
Message-Id: <970516080614_88054208@emout16.mail.aol.com>
Errors-To: madole@mills.edu
Reply-To: tuning@eartha.mills.edu
Originator: tuning@eartha.mills.edu
Sender: tuning@eartha.mills.edu