Re: Microtonal guitar/synth questions...

Hi there, Page Wizard.

I htink you have two things in mind - playing in various
scales that use pure ratios etc - this is easy using midi
relaying, or adaptive tuning. That is hard, as the notes of
a chord are not played exactly simultaneously, and it is hard
for a program to guess in advance what the remaining notes are.

Pure ratio scales can be fun to explore' limited number of
chords that are pure, but you learn where they are, and
interesting to develop music with that "universe". The hexanies
and other lattice scales can have large numbers of pure chords
and can be quite fascinating, though hard to navigate for
composing or improvisation on first encounter I find. Well
worthwhile getting to know one of those and writing something in
it.

For adaptive tuning, John deLaubenfels explores leisure retuning,
where one does a complete performance, and his program goes
through the whole thing and retunes it all. That sidesteps the
problem of not being able to anticipate the notes before they
are played.

Other approach is to have a way that one can specify oneself
what the root of the chord is. This is prob. quite easy,
e.g. with a separate manual, and you press appropriate key
for the root in it.

We are actually discussing all this over in crazy-music at present.

I'm author of fractal tune smithy which can be used for midi relaying.

See the "compositinal methods" thread, e.g. my post:

/crazy_music/topicId_965.html#974

part of a long thread with Carl Lumma and I discussing these ideas,
before and after that, and see if any is of interest to you.

Also earlier, been discussing adaptive tuning ideas with JdL.

I'm interested in extending the midi relaying capabilities of FTS.

Maybe you'd like to join the discussion.

crazy_music and makemicromusic are the groups really focussed on
how to actual make music for all the lattices and scales etc
discussed on the tuning group.

Just a first impression, getting a program to guess what notes
are likely to be played next is probably hard, if that is needed
- it is the sort of general area where computers are not so good
at as humans. But there are powerful techniques around in computing
that can make some progress even with intractable seeming problems like
this. Viz improvements in speech recognition, automatic translation
of languages etc. Okay automatic translations are not perfect, but
pretty amazing when one considers the computer has no knowledge of what any of
the words actually "mean" - couldn't point to any real world object
and says "this is a ...". In the same way, the midi retuning software
doesn't actually "understand" any of the music it is retuning, and
if that is what is required, it will be well out of its depth. But
some progress is possible and the number of mistakes it makes could
be reduced.

Only other things one could try are, to just allow glitches where
notes get retuned after they are first played, sliding up or down
in pitch, or to introduce a delay of say half a second between
playing a note and actually hearing it, to give program a chance
to identify all the notes of a chord and retune them accordingly.
One could prob. learn to play with such a delay, as it is
similar to non electrically controlled acoustic organs.

Robert