>
> ---------- Forwarded message ----------
> From: Mason Hock <[email protected]>
> To: lilypond-user <[email protected]>
>
> Subject: slurs between pairs of notes horizontally close and vertically
> distant
> I'm working on a score with many pairs of notes a sixth or seventh apart
> slurred together. When the stem of the second note of a pair is down, the
> slur intersects the accidental. If I increase
> 'Slur.details.accidental-collision'
> the slur avoids the accidental, but is not as well-shaped. For most pairs
> Lilypond acheives a good compromise, but sometimes the slur is quite
> distant from the notes and the endpoints of the slur are close to the same
> height, making it unclear that the curve represents a slur between the two
> pitches.
>
> It seems that the bad slurs occur when there is not a lot of horizontal
> space between the two notes. In the example below, the slur between the
> half notes in the second bar looks much better than the slur between the
> same two pitches as eighth notes in the first bar. Even the third and
> fourth slurs look much better than the first and second because the
> accidentals add just a tiny bit more horizontal space.
>
> From reading this <http://lilypondblog.org/2013/
> 11/engraving-challenges-slurs-and-ties> it sounds like slurs are very
> difficult to shape well automatically. It's impressive that Lilypond does
> this well before manual tweaking. However, this issue occurs so many times
> in this score that it would take way too long for me to fix each one
> manually with \shape.
>
> Since Lilypond really only seems to have trouble when the notes are very
> close to each other horizontally, perhaps a way to fix the slurs globally
> would be to set a minimum horizontal space between notes. However,
> Lilypond's horizontal spacing is not the issue and I'd rather not override
> something that it is doing well. Also, I expect that the horizontal space
> needed to fix the slur depends on the vertical distance and stem direction
> of the first note
>
> I would prefer a solution that only affects the slurs. Perhaps the issue
> would be fixed by allowing greater curvature for slurs. For example, if the
> first slur began close to the staff, looped all the way around the flat,
> then came back down and ended close to the second notehead, it would look a
> little odd but would be a large improvement. Some of the more adequate
> slurs might also be improved by this, and I don't see any that it would
> break. I would have no idea how to implement this though.
>
> Does anyone have any suggestions?
>
> Thanks,
>
> Mason
>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
> \version "2.18.2"
> \language "english"
>
> % slurs are shaped well but intersect accidentals
> \relative c' {
> \accidentalStyle neo-modern
> c'8( af') cqs,( aqf') cs,( a') ctqs,( aqs') | c,2( af') |
> \tuplet 3/2 {f,4( e') fqs,(} \tuplet 3/2 {eqs') gf,( f')} |
> g,( ef') gqs,8( eqf') gs,( e') |
> }
>
> % slurs avoid accidentals but are not well-shaped
> \relative c' {
> \accidentalStyle neo-modern
> \override Slur.details.accidental-collision = #30
> c'8( af') cqs,( aqf') cs,( a') ctqs,( aqs') | c,2( af') |
> \tuplet 3/2 {f,4( e') fqs,(} \tuplet 3/2 {eqs') gf,( f')} |
> g,( ef') gqs,8( eqf') gs,( e') |
> }
>
Here is an approach that yields something hopefully more acceptable.
I suspect this may cause some side effects, exaggerating other slurs that
were ok before, since it is a global adjustment.
\score {
\relative c' {
\accidentalStyle neo-modern
<>^"9"
c'8( af') cqs,( aqf') cs,( a') ctqs,( aqs') | c,2( af') |
\tuplet 3/2 {f,4( e') fqs,(} \tuplet 3/2 {eqs') gf,( f')} |
g,( ef') gqs,8( eqf') gs,( e') |
}
\layout {
\context {
\Voice
\override Slur.height-limit = 9
\override Slur.ratio = 0.5
}
}
}
HTH,
David Elaine Alt
415 . 341 .4954 "*Confusion is
highly underrated*"
[email protected]
self-immolation.info
skype: flaming_hakama
Producer ~ Composer ~ Instrumentalist
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
_______________________________________________
lilypond-user mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/lilypond-user
