Re: Lilypond benchmarking

[David Kastrup]

Keith OHara <address@hidden> writes:

> Phil Holmes <mail <at> philholmes.net> writes:
>
>> >> > With ragged-bottom set and annotate spacing, on one of my scores 17.19
>> >> > shows 49.07 space left and the next system having an extent of 44.7.
>
> The "space left" shown is after the lines have been laid out and
> skylines drawn.  While choosing page-breaks (before line-breaking 
> has been done)

Page breaks before line-breaking?  Unlikely.

> LilyPond is conservatively using the extents, which would have filled
> a bit more of the page.

I think we need to change that, or the net result of the skyline patches
will be _more_ pages (as they call for somewhat more padding).

> (It also seems that "space left" imagines we can overlap Copyright in
> the footer.)

That sounds like a flowery circumscription of "the page estimate is
wrong".

> The next system extent is (-44.70, 4.34) which is a height of 49.04,
> Given that the first-page extents plus copyright line would leave an
> estimated (at page-breaking time) space left somewhat less than 49
> spaces, I see why LilyPond put the next system on the next page.
>
> If I revert the 'clarify' patch, 
>  http://codereview.appspot.com/13768045
> that extent goes down to (-44.70, 1.31)  There are three 
> removed-as-empty staves at the top of that system, so it looks like 
> the 'clarified' code accidentally includes 1.01 staff-space per
> removed line.   
>
> I can't see the problem, so when I get time I'll step through the
> code to find where things go different, and if I don't figure it
> out before going out of town I'll just revert the 'clarify' patch.

I'll take a look as well.

-- 
David Kastrup