On 15/02/11 09:28, ?????? - Hu Haipeng wrote:
Hello,
I'm almost near the end of my second work (still about 80-100 bars not
written). The day before yesterday, when I was at bar 362, Lily could
still compile it within 2 hours. But last night, I left my computer
running it, and it was still hanging there when I got up this morning,
thus about 6 hours past! I am still using my old machine with 768M
RAM. Could anyone compile it using a faster machine? I must ensure
whether the problem is caused by the RAM or Lilypond or my own files.
The following is the link. Simply run dance.ly. But before that,
please find and remove %{ and %} which I use to generate midi only. I
included a midi file, so you will know what I have written.
http://www.darajan2.com/douban/Haipeng/Dance%20of%20heart.zip
I downloaded the zip and gave it a try (Ubuntu amd64 on a machine with a
core i7 CPU and 12Gb RAM; Lilypond 2.13.49). First attempt ran for about
five seconds, producing dozens and dozens of midi error msgs like so:
programming error: Impossible or ambiguous (de)crescendo in MIDI.
continuing, cross fingers
MIDI output to `dance.midi'...
programming error: Going back in MIDI time.
continuing, cross fingers
and then stopped with a message saying "Compilation successfully
completed", though only a midi file was produced, no PDF.
I commented out the \midi { } in the score block and ran another
compilation. When I checked after a few minutes, System Monitor showed
that Lilypond was using 1.1Gb of RAM, which goes a long way to
explaining why your machine with 768Mb RAM can't build it within any
reasonable timeframe, as continual swapping to disk would be making the
build very slow.
The score took 705 seconds (eleven and three quarter minutes) to
complete building and produce a PDF. So I'd say the solution is more RAM.
Nick
_______________________________________________
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user
