I've just contributed a transcription for harpsichord of Pachelbel's Canon to the archive. (http://www.icking-music-archive.org/scores/pachelbel/canon.pdf). I made this several years ago when I played a wedding gig and had a request to include the piece. I did find a solo keyboard version on IMSLP but it is heavily arranged and doesn't maintain the canonic structure. I wanted a cleaner, stricter version.
It raised some interesting transcription and typesetting issues. I thought others might be interested in how I attacked them. For playability and clarity, I decided to include only two treble voices (instead of the original three) plus the bass line. Both treble voices are always in the treble clef. I decided to include markings for when I would play the lower of the two voices with my left hand, and had to concoct a couple of TeX macros using \hrule to draw the right-angle symbols for that. (see the PMX source). I think using "l.h." and "r.h." would have been more intrusive and sometimes difficult to make clear because of restrictions on placement. The macros have two arguments, the level of the short horizontal leg and the length of the vertical leg. I suppose I should include them in pmx.tex. I assumed the player would guess that the l.h. symbol would hold for every lower note in the treble staff until cancelled out by an r.h. symbol. Except for a very few spots where 9ths and minor 10ths must be played with one hand, it can all be done with reaches of an octave or less. For the most part, I used upper stems for one voice for the duration of each 2-bar repetition, and lower stems for the other, although I didn't necessarily keep the same stem direction for the same voice when moving from one repetition to the next. Of course this still leads to lots of places where up-stemmed notes fall below simultaneous down-stemmed ones. In those cases I usually offset the upper note to the right by 0.2 notehead widths with either X.2S or, for a string on shifted notes, X.2: ... X: . I think the small compromise in readability is offset by the value of having a clear visual record of where each voice goes. I used two different mechanisms to deal with places where the same note occurs in two voices. If the two notes are attacked at the same time, I surrounded one by parentheses, using o( and o) on the one that should not be struck. On the other hand, if a note must be struck while another note at the same pitch is sounding from earlier, I terminated the holding note and inserted a rest as a visual cue to get that finger out of the way. A final comment on vertical rest placement: with the option AK in PMX, I didn't need to manually adjust the vertical positions of any rests at all; PMX did all that work. For example see bars 23-28. --Don Simons ------------------------------- [email protected] mailing list If you want to unsubscribe or look at the archives, go to http://tug.org/mailman/listinfo/tex-music
