Re: regression? [was Re: BeatLength and BeatGrouping]

[Carl Sorensen]

On 9/27/09 10:23 AM, "luis jure" <address@hidden> wrote:

> 
> 
> on 2009-09-27 at 08:02 Carl Sorensen wrote:
> 
>> The autobeaming behavior (and the documentation) have been fixed in
>> 2.13.4.
> 
> i find a regression in 2.13.4 with respect to 2.13.3.
> 
> i recently updated to 2.13.3 and i was very happy to see that the
> automatic placing of beams had improved greatly compared to previous
> versions. but in 2.13.4 some things seem to have regressed.
> 
> see for example the output of this code in both versions:
> 
> \relative c'' {
>         d16 d8. d8 d d16 d8. d8 d |
>         r16 d8. d8 d r16 d8. d8 d |
>         r16 d8.~ d8 d r16 d8.~ d8 d | }
> 
> 
> in 2.13.4 the dotted eights are beamed with the following eights if (and
> only if) preceded by a rest. this doesn't look well to me, specially
> when the notes are tied.
> 
> in 2.13.3 this was correctly sorted out. or i should say, the way _i_
> think it should be sorted out... :-) any other opinions?

We've had a serious discussion about whether eighth note beams should end on
every beat or only on beats 2 and 4.

We came down on the side of ending eighth note beams on 2 and 4, because it
seemed to fit Stone's rules.

http://article.gmane.org/gmane.comp.gnu.lilypond.devel/23515


You can return to the 2.13.3 behavior by adding this line to your file:

\overrideBeamSettings #'Score #'(4 . 4) #'end #'((* . (1 1 1 1))
                                                 (32 . (4 4 4 4 4 4 4 4)))

This will cause all beams to break at the beat, except 32 beams will break
at 1/8 notes.

HTH,

Carl