lilypond-user
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: midi2ly


From: François Pinard
Subject: Re: midi2ly
Date: Tue, 12 Apr 2005 18:37:22 -0400
User-agent: Mutt/1.4i

[Antonio PALAMA']

> [...] a properly encoded midifile contains more information than a
> printed score.  A printed score contains all the information necessary
> to a musician to play the music.  A midi file contains the actual
> performance of the musician.

A score encodes a musical intent, while a MIDI file encodes a
performance.  The author intent, expressed in a score, is a precious
source of indications, out of which musicians may deliver myriads of
different _and_ interesting renderings.  This is extremely rich.

No doubt that a particular performance may contain a lot of information.
You may go way further than MIDI, which is so fuzzy if you consider all
MIDI synthesizers around, and encode 44100 positions of the speaker
membrane for each second of the play.  But then, we are far away from
the author intent, and in fact, one may not fully recover nor induce the
original author intent from a particular rendering.

The amount of raw information does not matter so much here.  But the
amount of genuine musical information does.  And going from a score to
MIDI, we surely gain a lot of raw information, but on average, we loose
part of that genuine musical information designed by the composer.


P.S. - I also agree that some musicians prefer to perform (or improvise)
than to formalize their intent, or just do not have enough courage or
introspection abilities to do such formalization.  They might produce
quite interesting music nevertheless.  In such cases, granted, MIDI is
better than nothing at all.

-- 
François Pinard   http://pinard.progiciels-bpi.ca




reply via email to

[Prev in Thread] Current Thread [Next in Thread]