Style notes for mig-transcribers
--------------------------------

    The purpose of this little file is to provide new transcribers with a
number of advices that will make their coding easier and better. It will
maybe grow as time goes by, enriched by users' commentaries and music files.
The advices are not sorted, since there is no relation between them.

    - first of all, master rule: "if it sounds good, it IS good"

    - polyphony (two simultaneous "melodies" with the same rythm but
different pitches) sounds OK in Mig only if both voices are not too close;
if they are in the same octave and less than 4 or 5 half-tones away from
each other, you should consider lowering or raising one of them

    - avoid playing identical notes at the same time, except when roles of
both voices are clearly different (for example, melody + bass, or melody +
more rapid or slower accompaniment); it usually does no sound well; instead,
lower one of the notes (even in the middle of a melody, just for the
problematic note), and check if it sounds better

    - do not write two quavers if it is a crotchet; that is, do not write
(in $$-> format) "12abCCDE" if it really is "12ab11C12DE", or (in $->M
format) "12aCaDaDaE" if it really is "12aC11aD12aE", just to save nibbles or
key-strokes; it could be shorter in the source file but

        1) it is longer in the compiled file (7.5 additional bytes per
            additional note)

	2) the transition from one pair of notes to an identical pair of
	    notes is not always clean; what is clean is the transition of
	    one note to an identical note, if the other voice changes,
	    because this latter change will mask the transition (if
	    performed skillfully, as Mig always does); thus, if "you write
	    two quavers when it is a crotchet", you will get "unclean" music

    - a little thing about volume; if you ever worry about relative loudness
(you should, you can achieve amazing results if you do), you probably would
like to know the following facts:

        * what matters is the volume plus four; i.e., "P6a" ($$->syntax)
	    sounds twice as loud as "P1a", because 6+4=2*(1+4)

        * an octave higher gives twice as loud a sound; i.e., "O5P6a" sounds
	    twice as loud as "O4P6a"; or, using the previous rule, "O5P1a"
	    sounds as loud as "O4P6a"

        * the two previous rules are not true; they are only theory; it is
	    good to know them, but be aware that they will be accurate only
	    for not too high volumes, not too low pitches and not too high
	    pitches

    - choose the right music; when you have listened to a few mig files, and
made some attempts at transcribing music, you will be able to tell before
even beginning a transcription whether it will sound great or not so great



    A last word: please help me to enlarge this set of advices, by giving me
your advices and your comments; thanks.
