On Tuesday, 22 August 2000, Peter Schaffter writes:
> I tried removing all but one or two property calls from the file
> that was taking so long to process (see my previous post), and
> it seems that it is indeed the way lily handles \property calls
> that's eating up so much memory. It only took about ten minutes
> to process, instead of 80 minutes.
Good thinking! Luckily, this is just where we think (or actually
Han-Wen thought) some improvements could be made, in terms of memory
footprint.
Two more questions: what is the amount of memory that Lily consumes
for both cases (plug: you can use my grand proc-time tool:
http://appel.dyndns.org/fred/software/proc-time.c)?
Also, what happens if you remove all slurs (or remove the slur
engraver)?
[snip declaration of love to Lily]
> I'd love to post an example of the score I'm working on, to
> demonstrate my problem, but to be honest, I have yet to generate
> the dvi file of a fully corrected version. Maybe in a few
> months, when my hard disk stops thrashing (if it doesn't wear
> out from the overwork), I'll be able to.
Another, really silly option, is for you to tear up the file into
say four pieces. See where the line breaks are now, or where you
want them, and use a preprocessor to comment-out 3/4 of the score.
> Meantime, a couple of problems in 1.3.75, which perhaps have
> been corrected.
> 1. Tenuto marks, when they're directly over or under a stem,
> should be centred on the stem, not the note (as they
> currently are in 1.3.75).
Noted.
> 2. The tenuto marks currently have the same weight as ledger lines.
> They should be heavier.
I'll have to look into this. It's customisable anyway. I've seen
them lighter and heavier than ledger lines.
> 3. Tenuto marks, when they're under notes that fall on ledger
> lines, are not equidistant from all the notes. They're
> closer to notes that fall _on_ the ledger line, and further
> away from notes that fall in the spaces.
Noted.
> 4. 1.3.75's treatment of slurs that carry over from the end of
> one line to the next is not good. The general motion of the
> slur (i.e. up or down) is not respected, and the slurs clash
> with the clefs.
>
> Lastly, what does the error message "Degenerate bow: infinite
> steepness required" mean?
The slurs are still broken.
Greetings,
Jan.
--
Jan Nieuwenhuizen <[EMAIL PROTECTED]> | GNU LilyPond - The music typesetter
http://www.xs4all.nl/~jantien | http://www.lilypond.org
