Comment #11 on issue 1661 by [email protected]: MIDI dynamics take effect
one note too late
http://code.google.com/p/lilypond/issues/detail?id=1661
Confirmining. I was stating independent facts back in comment 4 :
This is one of Heikki's MIDI patches; it fixes the original problem
completely.
The original problem is a regression; for practical purposes, ver 2.12
implemented dynamics for proof-hearing purposes, while ver 2.14-current do
not.
The suggestion in comment 1 is a request for enhancement (automatically
shifting dynamics to a preceding grace) that Heikki did not include in his
patch. I don't want that enhancement myself, but anybody who does can post
it as an enhancement request.