Re: subdivideBeams broken?

[Carl D. Sorensen]

On 6/21/09 3:27 PM, "Hans Aberg" <address@hidden> wrote:

> Yes, beatLength will do a (2+2)+(2+1) beaming. Though this is one
> possible beaming for what I am writing now, my problem is that I have
> bunch of different meters. For example, I may want (2+2)+3,
> (2+2)+3+(2+2) and so on.
> 
> I have looked at the stuff you describe above ("Sub-dividing beams",
> page 60/70 in the manual). It seems that beatLength assumes that
> subdivisions should happen on multiples of this time values. This does
> not work with a meter beaming like 3+(2+2) or (2+1)+(2+2). See code
> below - it seems that this 3 causes problems.

Yes, this is a known limitation of the beam subdividing.  It is only
implemented for beatLength adjustments.

Personally, I don't like using beatLength for beam subdivision.  To me,
beatLength should have one meaning only -- the numerator of the time
signature.

 
> The auto-beaming model is perhaps too crude for subbeaming all these
> meters - in some, a trick might do.

It's not the auto-beaming model, but the beam-subdividing model that is too
crude.  And that is unfortunate, because I know of no way to do a manual
beam subdivision.

But there has been a proposal floated to fix this, so maybe it will be
improved....

Carl