Re: Issue 190 in project lilypond

[Mats Bengtsson]

Han-Wen Nienhuys wrote:
> Mats Bengtsson escreveu:
>   
> > I just tried the example again with the latest git version and
> > the beam is slanted upwards (not much but still clearly visible
> > if you look carefully).
> >
> > I have attached a png with debug-beam-scoring and another one
> > where I also set inspect-quants. The corresponding input file is repeated
> > here for completeness.
> >
> > \version "2.11.7"
> > \layout{
> > %  line-width = 5\cm % Gives expected beam slanted downwards
> > line-width = 10\cm % Gives unexpected beam slanted upwards
> > debug-beam-scoring = ##t
> > }
> > \relative c'''{
> > \time 2/4
> > \override Beam #'inspect-quants = #'(1 . 2)
> >     
>
> for 1 . 2 you have substitute the positions of what you'd want to see.
>   
OK! Here comes a new test case and the corresponding output.
No matter if you prefer the 2nd or 3rd of my example, I hope everybody
agrees that the inverted slope of the 1st example (i.e. the current default)
is worse.
\version "2.11.7"
\layout{
line-width = 10\cm % Gives unexpected beam slanted upwards
debug-beam-scoring = ##t
}

% Default layout with inverted beam slant
\relative c'''{
\time 2/4
fis16 dis b ais cis4
}

% Better alternative, similar to the beam position when line-width=5\cm
\relative c'''{
\time 2/4
\override Beam #'inspect-quants = #'(0.3 . 0)
fis16 dis b ais cis4
}

% Another alternative, with somewhat steeper slope.
% Should also score higher than the current default, IMHO.
\relative c'''{
\time 2/4
\override Beam #'inspect-quants = #'(0.6 . 0)
fis16 dis b ais cis4
}

  /Mats