Re: Shortening eighth note beamed stems and other typesetting questions

[Thomas Morley]

Am Di., 4. Mai 2021 um 17:34 Uhr schrieb Knute Snortum <ksnortum@gmail.com>:
>
> Hi, everyone.  This will probably be a long post.  I hope you will bear with 
> me.
>
> I have two questions: one as experts in LilyPond and one that just has
> to do with typesetting, but they are related.
>
> First, I am trying to shorten the stems of beamed eighth notes.  I
> found that Stem.details.beamed-lengths could do that, but it seems to
> have minimums that it won't go past. Consider this snippet:
>
> %%% Start
> \version "2.22.1"
>
> \relative {
>   \override Stem.details.beamed-lengths = #'(2)
>   c''8 c c c
>   \override Stem.details.beamed-lengths = #'(3)
>   c b a g |
>   \override Stem.details.beamed-lengths = #'(3)
>   c a f d
> }
> %%% End
>
> I can't get the repeated "c" eighth notes lower than 2 or the
> descending eighth notes lower than 3. Is there a way to get past these
> minimums? Or am I doing something wrong?
>
> Now the typesetting question: I have a piano piece that has several
> voices and there is a run of descending eighth notes inside of other
> voices that gets really cramped. Attached is a screenshot of how some
> older editions dealt with it (old-edition-solution.png). It requires
> shortening the eighth note beam down to almost nothing.
>
> Here is how the two measures look without any intervention
> (eighth-note-stems-without-intervention.ly produces
> lilypond-solution.png). It's an interesting solution, but I don't
> think it will work.
>
> And here is my solution (eighth-note-stems-my-solution.ly produces
> my-solution.png). Is it readable? Should I try to make it look like
> the old edition? If so, how?
>
> Thanks for reading this to the end.
>
> --
> Knute Snortum

For completeness:

The override for Stem.details.beamed-lengths is not always sufficient,
because Stem.details.beamed-extreme-minimum-free-lengths and
Stem.details.beamed-minimum-free-lengths provide some minimums.
In own scores I had set them to zero, adjusted
Stem.details.beamed-lengths to taste and used Beam.positions for
fine-tuning.

\relative {
  c''8 c c c
  \override Stem.details.beamed-lengths = #'(1)
  \override Stem.details.beamed-extreme-minimum-free-lengths = #'(0)
  \override Stem.details.beamed-minimum-free-lengths = #'(0)

  \override Stem.no-stem-extend = ##t
  c8 c c c
  c b a g
  \once \override Beam.positions = #'(1.8 . 0.5)
  c b a g
  c a f d
  \once \override Beam.positions = #'(1.8 . -1.1)
  c' a f d
}

As soon as 16th happen other values are preferable, thus Carl's method
looks superior (not tested, though).

Cheers,
  Harm