Replies to John and Paul about 19-tet guitar piece

John and Paul (nothing for George or Ringo)

Thanks for your comments about the 19-tet piece.

If I may summarize, you suggested that it would be
good to create some "companion"
pieces for the purposes of comparing:

(a) 19-tet piece with harmonic timbres
(b) 19-tet piece with spectrally mapped timbres (this is the "real" one)
(c) piece changed to 12-tet with harmonic timbres, and
(d) piece changed to 12-tet with timbres spectrally mapped (to 19-tet)

I've been trying to figure out exactly how to
do these. While they seem straightforward, a bit of
thought shows that there are problems in
making a "fair" comparison.

As I mentioned in an earlier post, just listening to the
raw sound samples, the main difference appears to be a slight
"phasing" kind of effect (as you well know, the mapping to 19-tet
is quite benign compared to many). There was no audible pitch shift,
and the timbre didnt change much, except for the phasing.

Now, when I mixed down the piece (b), I added some reverb and
a small bit of phase shifting (normal things for a "guitar")
- of course, I chose the amounts so that it
sounded good to me... so - problem #1: how much reverb and phasing
should be added to the version (a)? If I leave it the same, then
very likely its the "wrong" amount... that seems like its crippling
version (a), yet if I change it, then we've lost experimental control!

The problems are far worse for (c) and (d). Here, we must decide
about the slides... I adjusted the portamento (for instance)
so that you got a smooth glide between adjacent 19-tet steps
in roughly the time span over which they were being played...
when mapping to 12-tet, should this glide time be adjusted
to the time required for 12-tet steps? If no, then the 12-tet piece
is at a disadvantage since the "correct compositional decision"
has not been taken. However, if we do change it, then we have lost
experimental control!

Because the effects in this 19-tet case are subtle (the "phasing")
Im not sure that this is worth it... any differences are likely to be
swamped by the various arbitrary trade-offs that are necessary.

On the other hand, John had mentioned that he would like to
try some stuff with the MIDI file - Id be happy to share this,
to see what kinds of things you can come up with...

As a final note, I have done comparisons very much like (a)-(d) on
the Turquoise Dabo Girl on the CD that cames with TTSS. While
track 26 plays the piece with the mapped timbres in 11-tet, track 27
plays (the first 16 bars) using "the same" (i.e., the unmapped)
harmonic timbres. While problems like the ones mentioned above
also occurred here, the major effect is the "out-of-timbre"
perception, and this dominates minor things like reverb.
If you would like, I can put this on the web...

Bill Sethares

Hi, Bill

The last thing I want to do is throw distracting challenges at you in
mapping your 19-tET piece to other tunings and timbres. I appreciate
the subtleties of your concerns.

(Of course, if you have a MIDI file at hand, I'd be glad to play with
it)

I'd much rather pick your brain regarding preparing altered timbres for
one or more complete GM voice sets. It seems as if a fair amount of
tinkering with the overtones is almost inaudible to the ear on
individual notes, while making a huge difference to the implicit "JI"
tuning, as you and Margo (and others) have demonstrated.

How hard would it be for a guy like me to acquire and use the tools that
would be required for such a project? Is it even feasible? As you
know, I'm a long way from being the expert you are in such matters.

Any crumbs of knowledge, pearls of wisdom, would be appreciated.

JdL