11 from 19 and tuning

How I make micro music:

I use sibelius and a Roland Sound Canvas SC88. In Sibelius you input pitch
bends as two numbers, the the first of which is 0, as the fine pitch bend does
not work with my Sound Canvas (or sib ignores any numbers you put in there)

the notation is ~B0,nn

where nn is in the range -128 to 127. So if you set the maxx range of the pitch
bend to 1 semitone up and down, you get 2 semitones divided into 255 parts,
which is not a brilliant accuracy. but unless anyone has better experience with
Sib v.3 and microtones and Roland Soundcanvases, to achieve better
intonation results, it's all I can do. Deviations from *exact* 19equal are
products of the technology not impure compositional thinking.

I anyone can produce a rendition using the exact implied tuning, I'd be
interested to hear also. To suggest that I was using a different tuning by
analysing an electronic performance is incorrect - and in any case I suspect
any real acoustic performance on two real recorders will have deviations far
worse than my electronic ones, so the discussion vis a vis accuracy is largely
moot.

Still, I'm glad people like what I've written.

Mark

--- In MakeMicroMusic@yahoogroups.com, "microtonalist" <mark.gould@a...> wrote:
> How I make micro music:
>
> I use sibelius and a Roland Sound Canvas SC88. In Sibelius you input pitch
> bends as two numbers, the the first of which is 0, as the fine pitch bend does
> not work with my Sound Canvas (or sib ignores any numbers you put in there)
>
> the notation is ~B0,nn
>

I can only speak from experience from sibelius 2. The pitch bend notation is the
same there, but I have not used it extensively. I have noticed, though, that Sibelius
has a plugin for "quartertone playback" that looks for quartertone accidentals (which
sibelius supports but does not automatically play back) and sticks in the necessary
pitch bend. I was thinking, couldn't someone (dangit I'm probably volunteering
again) write plugins that would implement pitch bends in other tunings? 17, 19 and
even 31 would be easy; you could even do unequal tunings as long as the notation
you want is limited to the normal plus quartertone accidentals. Anything else and I
think there would be trouble.

The programming language is Sibelius' own, they're apparently open to including
plugins if they like what you send, and they say they'll even pay money for good ones!
(Maybe I don't need a summer job...) And what better way to infiltrate mainstream
music making than with "Let's see what your piece sounds like in 19 equal..." (okay
I'll admit there are better ways but whatever it takes to get people thinking...)

--- In MakeMicroMusic@yahoogroups.com, "microtonalist"
<mark.gould@a...> wrote:

> where nn is in the range -128 to 127. So if you set the maxx range
of the pitch
> bend to 1 semitone up and down, you get 2 semitones divided into 255
parts,
> which is not a brilliant accuracy.

Ah ha! Your 11-note scale in terms of midi note numbers and pitch bends is

[[61,6528], [62,8576], [63,10752],[65, 9856], [64, 7808], [66,10972],
[68,6016], [69,8192], [70,5120], [71,7296], [72, 9472]]

If you look at the second note of each pair, they are all divisible by
128 = 2^7 except for [66, 10972]. I still don't get why you can't come
a lot closer to the fifth of 19 equal, however; one of your notes is
[71, 7296], and that is as close as you can come using pitch bends
divisible by 128 (course pitch bends only, or 1536 equal) to a 19-et
fifth over [64, 8192]; this would be great except that [64, 8192]
isn't in your scale.

but unless anyone has better experience with
> Sib v.3 and microtones and Roland Soundcanvases, to achieve better
> intonation results, it's all I can do. Deviations from *exact*
19equal are
> products of the technology not impure compositional thinking.

> I anyone can produce a rendition using the exact implied tuning, I'd be
> interested to hear also.

I could change it to 19 equal easily enough, and I suspect if I do it
will, after all, turn out to be semisixths.

To suggest that I was using a different tuning by
> analysing an electronic performance is incorrect - and in any case I
suspect
> any real acoustic performance on two real recorders will have
deviations far
> worse than my electronic ones, so the discussion vis a vis accuracy
is largely
> moot.

I'm simply reporting on the tuning as it exists in your midi file.

--- In MakeMicroMusic@yahoogroups.com, "Gene Ward Smith"
<gwsmith@s...> wrote:

> If you look at the second note of each pair, they are all divisible by
> 128 = 2^7 except for [66, 10972].

And that turns out to be a copying mistake. All the pitch bends are
coarse; divisible by 128.