Comment #11 on issue 1776 by lemniska...@gmail.com: Doc: NR - Polymetric
Notation \compoundMeter isn't documented
http://code.google.com/p/lilypond/issues/detail?id=1776
James,
i hoped to answer sooner but i had to fight problems with grand-fixcc.
2011/7/31 Comment #9 on issue 1776 by pkx166h:
However, because I am curious (and don't a thing about this special
notation) can you tell me why
\relative c' {
\compoundMeter #'((1 4) (3 8 4))
\repeat unfold 5 c8 \repeat unfold 10 c16
}
This one is perfectly ok and compiles for me.
\relative c' {
\compoundMeter #'((2) (3 8 4))
\repeat unfold 5 c8 \repeat unfold 10 c16
}
You cannot have only one number in parentheses.
Fails? and why I get a different error for each case?
Id kind of expect
\compoundMeter #'((2) (3 8 4))
to be equiv of 2
3 + 8 + 4
Is this not allowed in music?
Or at the very least 2 + 2 + 2
3 8 4
Maybe this is an enhancement?
Again as an example
\compoundMeter #'((1 4) (3 8 4))
could be the same as 1 + 4 + 4
3 8 4
I hope this makes sense?
I'm afraid not... It looks like you're reading the syntax in the wrong
way. I'll try to explain it in more detail (hoping that i understand it
correctly myself ;).
Time signature tells musicians how many beats are in the measure and what
is their length. Ordinary time signature is like a fraction: \time 3/4
says "there are three beats in the measure, each beat has a length of a
quarter note", \time 8/8 says "there are eight beats in the measure, each
beat has a length of an eight note" and so on: the numerator of the
fraction says how many beats are in the measure, and the denominator says
how long they are.
Now consider a piece written in 15/8 (this surely is a strange meter!). It
tells that there are 15 eight notes in the measure, but how the musician is
supposed to play it - how will he subdivide the measure? (for example, a
6/8 measure is divided into two groups, each consisting of 3 eight notes:
*one* two three *one* two three). What is the composer's intention in
case of 15/8? That's where \compoundMeter comes to rescue: one can write
\compoundMeter #'((4 4 4 3 8)) and the musician will know that he should
play it like this:
*one* two three four *one* two three four *one* two three four *one* two
three
writing \compoundMeter #'((3 3 3 3 3 8)) would mean to play it like this:
*one* two three *one* two three *one* two three *one* two three *one*
two three
Now imagine a meter in which beats have different length. For example,
first beat in the measure is a quarter note, and second beat is a half
note. How are you supposed to write this? The measure length is 1 quarter
+ 1 half = 3 quarter notes, but you cannot write \time 3/4 because that
means "three beats, each one quarter long". \compoundMeter #'((1 2 4))
wouldn't be accurate, too. You need to write this time signature as 1/4 +
1/2 - this is accomplished with \compoundMeter #'((1 4) (1 2)) .
So, using the "fraction" nomenclature, \compoundMeter #'((2 3 4)) produces
a single fraction (with the numerator consisting of two numbers: 2 and 3),
while \compoundMeter #'((3 8) (2 4)) produces two fractions (3/8 and 2/4).
Notice where the plus sign is located in each case. As you can see, each
parentheses inside \compoundMeter #'( ) corresponds to a separate fraction
in the final time signature; \compoundMeter #'((3 8) (1 2) (4 1 4)) will be
made of 3 fractions. If you write \compoundMeter #'((2) (3 8 4)), you are
telling Lily "do a complex meter consisting of two fractions: first
fraction will be made from number 2, second fraction will be made from
numbers 3, 8 and 4" - while one number (2) is not enough to form a
fraction: at least two numbers are necessary for a fraction!
Maybe it would be good to include some of this explanation in the patch?
I just want to know what the limitations there are and if these are
@knownissues or bugs or something else I need to document.
I thought i found one, but now it seems to works... It seemed that
\compoundMeter wouldn't accept irrational meters (for example
\compoundMeter #'((1 3) (1 4)) ), but now it works... Strange.
_______________________________________________
bug-lilypond mailing list
bug-lilypond@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-lilypond