Re: Stem length

[Hilary Snaden]

James Lowe wrote:
> > I've been trying these to lengthen some stems to avoid collisions...
> >
> > \override Stem #'(details beamed-lengths) = #'(5)
> > \override Stem #'length-fraction = #(magstep 3.5)
> >
> > ...but despite documentation to the contrary, non-integer values aren't
> > recognised, making it impossible to fine-tune the stem lengths. Is this
> > a known problem, or am I missing something?
> It might depend on your example, do you have something you can post here?

The piece is a song with a fairly complex piano accompaniment. This is 
the troublesome bar. With no tweaks, the beam of the inner semiquavers 
is rendered on top of the lower head of the concurrent bes's. If I set 
magstep to values from 3 to 3.3, the results are identical, with the 
beam of the inner semiquavers still almost touching the head. If I set 
magstep to values from 3.4 to 4, the results are identical, but with the 
spacing too large (at least for my fussy aesthetics).

If I use \override Stem #'(details beamed-lengths) = #'(5) instead, a 
value of 5 to 5.1 give results similar to magstep = 3~3.3 (as described 
above) 5.2 to 6 are similar to magstep = 3.4~4.

\version "2.12.2"

\score {
  \new PianoStaff
  <<
    \new Staff {
      \relative bes' { \clef treble \key des \major \time 2/4
      << { <bes ges des>4 <bes bes,> } \\
	 { s4 \override Stem #'length-fraction = #(magstep 3.5) aes='16( ges 
ees des) } >>
      } }
    \new Staff {
      \relative bes { \clef bass \key des \major \time 2/4
      <bes des,>4 ees,,16( bes' ees ges)
      } }
    >>
  }

I've just read Nick Payne's suggestion and found that this gives the 
result I was looking for:

\once \override Beam #'positions = #'(-5.5 . -6.5)

I'm posting my earlier findings in case they are of interest.

--
Hilary Snaden