Re: << \\ >> constructs and things that cross measures

[Trevor Bača]

On 8/25/07, Kieren MacMillan <address@hidden> wrote:
> Hi Ed:
>
> >    These voices are all separate from the voice that contains the
> > notes just
> >    outside the << \\ >> construct. This should be noted when making
> > changes at
> >    the voice level. This also means that slurs and ties cannot go
> > into or out
> >    of a << \\ >> construct.
>
> That's because of the \\ only -- that creates two new Voice contexts,
> each different from the main one that existed before the <<>>.
>
> What you *can* do is
>
>      a b c\< d << { a b\! c d } \new Voice { g g g g } >>
>
> and things will behave as desired -- the first {} is a continuation
> of the main Voice context, and only the \new {} is different/new. See
> "Instantiating new voices" in the docs for more info.

Yup, this is indeed very useful. Here's me discovering this construct
just a couple of days ago for double-voiced noteheads and stems that
cross over in piano music:

  http://lists.gnu.org/archive/html/lilypond-user/2007-08/msg00275.html

(See also Mats's more efficient follow-up post to that thread.)

The trick is in recognizing that there are two separate
hey-I-temporarily-want-polyphony  constructs. The first is << \\ >>
with the backslashes and the second is << >> without the backslashes.

And in the last couple of days I've realized something nice, too: if
you have a << >> polyphony construct in the middle of a tuplet, the
time-scaling of the tuplet transfers beautifully to all expressions
within the << >>. This means that you can, for example, pop a single
note inside some piano tuplet to a different staff somewhere and be
sure that the duration is the same as all other notes within the
bracket. Very useful. I'm costantly impressed that time-keeping works
as well as it does in LilyPond ...



-- 
Trevor Bača
address@hidden