Re: small staff massacre

[Werner LEMBERG]

[CVS 2004-01-22 10:01]

> > % . There should be a mechanism to give the staff size in a \with
> > %   {...}  block with a single command, covering all parameters.
> > %   Perhaps a new `set-staff-size' scheme function?
>
> Possibly - I don't consider this high priority.

I agree.  If two commands yield the expected results I can live with
it -- currently :-)

> > % 1. Note positions in chords are not computed correctly.  There
> > %    seems to be a problem to decide which is `up' and which is
> > %    `down'.
>
> Strange, cannot duplicate.

I still get this.  I use guile 1.6.4, compiled manually.  For both
lilypond and guile I've used gcc 2.95.3 (with libc.so.6).  Maybe a
rounding problem, relying on compiler-specific features?

> > %   3. The parameters to `extra-offset' aren't scaled properly.
>
> This is intentional. Extra-offset is a global mechanism that is
> supposed to work for arbitrary objects.
>
> We could change this, though.

While extra-offset now seems to give the correct results, the size of
dynamics is still different (see image).

> > % 10. I'm not sure whether the thickness of staff lines really
> > %     should be the same in both the bigger and the smaller staff.
> > %     Thicker than normal is correct, but that thick?  The problem
> > %     is that the thickness of beams must be smaller; this reduces
> > %     the contrast between beams and staff lines.
>
> Check out a good quality pocket score.

Hmm, I've just had a look into a Henle Urtext edition (violin concerto
No. 3 from Saint-Saëns, piano score), and there the staff lines of the
smaller violin staff *are* thinner.  But this is from 2002, and it
looks like a computer generated score.  An older Henle Urtext edition
(Beethoven violin sonatas, from 1978) indeed have equally thick staff
lines, but beams are thicker.  So I suggest to increase the beam
thickness slightly or to reduce the line thickness a bit.  [Some hours
later] After I've printed some pages with a 600dpi laser printer I now
think that thinner lines is the right solution.


    Werner