Phil Taylor wrote:
>John Walsh wrote:
>>[...]
>>M:3/8 L:1/8
>>F3|FFF|F3|FFF|F3|FFF|F3|FFF|F3|FFF|F3|FFF:|
>
>Surely the third voice is half as long again as the other two?
>If it were written like this it would be OK:
> [...]
>V:3
>M:3/8
>L:1/8
>F3|FFF|F3|FFF|F3|FFF|F3|FFF:|
No--there was a mistake in the part, but it wasn't that. (I just
forgot a double bar line.) The original piece had twelve bars of 3/8 time
in parallel with 6 of 2/4 and four of 3/4, with no tempo markings. (This
is copied directly from the MusixTeX docs, by the way.) Presumably the
musicians playing it would figure out the relative tempos by seeing which
notes lined up in the three staves. This was why I was thinking that some
sort of a tab stop before the double bar-lines (there should have been a
double bar ending the sixth measure in the third voice) would tell the
program to align the double bars, and give it a fighting chance to get the
proper notes aligned elsewhere. I have no idea what to tell a player
program to get it to play back as intended, tho.
(Sorry, James, I misunderstood your proposal: I was thinking about
tab stops at the time and thought you wanted to force voices to align at a
certain point, rather than to recognize the fact that they were aligned.)
I think it's a pretty safe prediction that whatever the V:
standard turns out to be, it'll be abused much as (and for the same good
reasons) the guitar chord mechanism is now: i.e. to write something that
abc wasn't explicitly designed to do, but *can be made to do*. (That's
called "hacking," isn't it? Maybe it'd be a good idea to design in little
hackability in the first place instead of worrying about how much abc
hacking's been done.)
I've been going thru the MusixTeX docs, looking for examples of
multi-staff music in order to see what has to be done to give abc2mtex
multi-staff capability, and I ran into couple of examples which had
isolated chords with notes of different lengths: e.g. three quarter notes
and one dotted quarter note. If abc requires all notes in a chord to have
the same duration (the 1.7.6 standard doesn't say that explicitly, but it
seems to be assumed) then you could hack the extra note with an additional
voice. If that's the only place in the piece that happens, you have to put
in a lot of invisible rests! (Or could you use V:sync right after that
note?) Even tho the extra note comes from an internal voice in the music,
this still strikes me as a slight abuse of the voice mechanism. And an
inelegant one, at that. Or...is this perhaps one of the things J-F
Moine's floating voice was designed for?
Cheers,
John Walsh
To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
