source file: m1447.txt
Date: Sun, 14 Jun 1998 10:29:49 -0500 (CDT)

Subject: Re: A pitch for making synthesizers tunable to an accuracy of at least 0.1 cent

From: mr88cet@texas.net (Gary Morrison)

>I believe that the difference between being able to set a
>pitch to within plus or minus 1 cent and being able to set that pitch to
>within plus or minus 0.1 cent could in some cases spell the difference between
>achieving a really powerful musical effect

   I'm inclined to give that the ol' "yes and no" response.  I agree very
definitely that accurate rendering of pitch can present some intriguing
differences in how harmonies are perceived.  Thus JI vs. 12TET vs. various
other temperaments.

   But I think it's also worth pointing out that, most real-world
instruments (keyboards being a very important exception) are a very
different story.  The vast majority of instruments are imprecise-pitch
instruments, meaning that their players have to constantly monitor their
pitch, often correcting flaws in the acoustic design of the instrument
itself.

   But the reality of the matter is that for most instruments, it's often
difficult to play normal-speed music with 5- to 10-cent accuracy much less
1-cent accuracy.  Not impossible, true, but difficult.  I often practice my
tenor saxophone with the tuning meter turned on constantly, and I'll look
down to see notes sometimes over a comma off the nominal 12TET pitch.
Granted, that error is probably what causes me to look down at the tuner.

   Or to present this in a more extreme light, I'm currently struggling
with some sort of really obnoxious tuning problem on my soprano saxophone.
Certain pitch ranges on the horn routinely come out (without correcting
them) nearly a quartertone off!  It's extremely difficult to correct the
pitch of an instrument that much, so even after correcting them, those
notes almost always end up much more than a comma off.  And this is
otherwise a high-quality instrument (Yamaha YSS-675, retailing for well
over $3000), and at least one historically very good repairman couldn't
find anything wrong with it!

   So my point is this:  I agree that there are intriguing subtleties for
sustained chords down in the 1-cent-resolution range.  But it's very
important to understand that people in general are used to VASTLY more
"slop" in pitch than that.  They don't routinely operate down in the
pitch-discrimination realm of single cents for normal-speed music.  I
therefore personally find more intrigue in tunings that produce entirely
different intervals (e.g., 9:7 instead of 5:4), or entirely different scale
structures (e.g., no meaningful approximation to a diatonic scale).

   To me personally, entirely new harmonic resources and melodic scale
structures are more interesting than one- to five-cent distinctions of the
temperament of a well-known scale structure.