> From: Trevor Daniels [mailto:[email protected]]
> Sent: Wednesday, 14 January 2009 19:25
> To: [email protected]; [email protected]
> Subject: Re: Hairpin #'minimum-length doesn't work if hairpin preceded
> by dynamicmarking
> 
> When applying minimum-length it seems the length of the
> hairpin is measured from the start of the note, not the start
> of the hairpin, i.e. the dynamic marking is included in the
> hairpin length.  You can see this if you make the length longer.
> 
Yes, I discovered that shortly after sending my message. I should have tried a 
few other values for the length first and it would have been apparent. However, 
it seems rather undesirable behaviour from the point of view of leading to 
unattractive output without manual overriding of some sort, plus when I ask a 
program to make a certain element X units long I don't expect it to decide to 
include the preceding element when calculating the length.

> The following change works better (with the missing
> back slashes added), although the pp still shortens the hairpin.
> 
>  \relative c'' {
>    \override Hairpin #'minimum-length = #5
>    c8\< c\! c\> c\!
>    c8\pp\< c\! c\> c\!
>  }
>
The problem with separating the markings for the crescendo and decrescendo 
hairpins is that they then won't line up horizontally if the pitch changes (eg 
c8\< d\! g\> a\!).
> 
> The following workround might suit you better, as the
> dynamic is now detached from the hairpin:
> 
>  \relative c'' {
>    \override Hairpin #'minimum-length = #5
>    c8\< c\! c\> c8*1/2\! s16\pp
>    c8\< c\! c\> c\!
>  }
> 
That works, but requires moving the pp if it falls at the beginning of a bar - 
add \time 2/4 to the example and it ends up in no mans land.

What I did eventually was to globally set \override Hairpin #'minimum-length = 
#5 and then add \once\override Hairpin #'minimum-length = #8 at the ten or so 
places where a dynamic marking was leading to a short stubby hairpin.

Nick



_______________________________________________
lilypond-user mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/lilypond-user

Reply via email to