Re: midstaff line = stem shortened?

[Carl Sorensen]

On 4/10/10 2:08 AM, "Mark Polesky" <address@hidden> wrote:

> Carl Sorensen wrote:
>> As I read Ross, the 1/4 staff-space adjustment is what one
>> would use if in the midst of a note cluster (perhaps
>> limited to a measure according to the text explaining
>> Example 1 and Example 2)  that contains the a, b, c, and
>> d, so we need to keep the stem progression moving.  If the
>> cluster only contains b, c, and d, then 2 1/2 works for
>> the b.  And if the cluster only contains a, b, and c, then
>> 3 works for all three.
>> 
>> But maybe the suggested compromise gets it so close, with
>> so little difficulty, that we should just implement it and
>> be done.
> 
> It certainly is tempting.  I've attached another file
> showing the Ross examples, comparing his suggested
> adjustments with my suggested compromise.  The Ross
> suggestions remain superior, but unquestionably trickier,
> since his adjustments depend on the neighboring pitches.
> But my compromise would still be an improvement on the
> current default, and wouldn't require any fancy parser
> gymnastics.

Well, I don't think that doing the Ross adjustments requires parser
gymnastics; it's engraver gymnastics that would be required.

> 
> So my ideal preference is to side with Ross, and if you want
> to cut your teeth on that, you'd totally make my day.  But
> if it's too tricky for now, I'd vote to do the compromise,
> since it's still better than what we have now; and the fancy
> Ross stems can always be coded at a later time.  I'll leave
> it up to you.

Actually, I'm not planning on attacking the stem problem.  My next plan
(once I get a spare minute, which might be 3 weeks or more) is to fix the
string-number on chords problem.  The one after that is to fix the
autobeaming system.  I think that both have higher priority than the stem
problem, at least in my opinion.

> 
> One note of caution: as I understand it, the discussion here
> *only* applies to non-flagged non-beamed stems.  Flags
> require a certain minimum stem height I think, so be sure
> not to jeopardize that.  Also, Kurt Stone makes a good point
> that multiple tremolo beams on a single stem may prohibit
> stems from being shortened; the stem should always extend
> slightly past the furthest tremolo beam.

There are lots of adjustments made in stem lengths to accommodate flags and
beams.    Ross has pages and pages of examples.  I've looked at the beam
quanting code in LilyPond (not that I've understood it, but I've looked at
it as I was getting started in autobeaming) and I'm pretty sure that
LilyPond gets it right at the moment.

Thanks,

Carl