Re: [tablatures] transposing tablature

["Patrick Schmidt" <p.l.schmidt@xxxxxx>]

-------- Original-Nachricht --------
> Datum: Sat, 9 Jan 2010 22:27:06 -0700
> Von: Carl Sorensen <c_sorensen@xxxxxxx>
> An: "tablatures@xxxxxxxxxxx" <tablatures@xxxxxxxxxxx>
> Betreff: Re: [tablatures] transposing tablature

> 
> 
> 
> On 1/9/10 12:17 PM, "Patrick Schmidt" <p.l.schmidt@xxxxxx> wrote:
> 
> > 
> > 
> > -------- Original-Nachricht --------
> >> Datum: Fri, 8 Jan 2010 20:47:50 -0700
> >> Von: Carl Sorensen <c_sorensen@xxxxxxx>
> >> An: Carl Sorensen <c_sorensen@xxxxxxx>, Patrick Schmidt
> <p.l.schmidt@xxxxxx>,
> >> "tablatures@xxxxxxxxxxx" <tablatures@xxxxxxxxxxx>
> >> Betreff: Re: [tablatures] transposing tablature
> > 
> >> 
> >> 
> >> 
> >> On 1/8/10 7:50 PM, "Carl Sorensen" <c_sorensen@xxxxxxx> wrote:
> >> 
> >>> 
> >>> 
> >>> 
> >>> 
> >>> On 1/8/10 12:05 PM, "Patrick Schmidt" <p.l.schmidt@xxxxxx> wrote:
> >>> 
> 
> >>>> You get an inverted 4-note d major (sixth) chord in first position.
> The
> >> root
> >>>> of the former c major chord got lost in transposition by default
> >> because the
> >>>> root of the resulting d major chord would be on the fifth fret of the
> >> fifth
> >>>> string. LilyPond tries to find the corresponding pitches in the first
> >> four
> >>>> frets of the guitar because the default value of minimumFret is zero.
> >> In my
> >>>> opinion this behaviour is wrong because not all the pitches of the
> new
> >> chord
> >>>> can be found in first position.
> >>> 
> >>> This is a bug.  Minimum fret should not default to zero.  It should
> >> default
> >>> to unset, and the code should be able to find the correct chord.
> >>> 
> >>> I had missed this problem before in your code.  I apologize for not
> >>> understanding it.
> >>> 
> > No need to apologize, it was my fault.
> >>> I think I can fix this bug.  I'll try to get to it in the next couple
> of
> >>> hours.
> > Thanks very much!
> >> 
> >> Patrick,
> >> 
> >> Can you add a FretBoards context and do some experimenting to see if
> the
> >> chords transpose properly in the FretBoards context?
> > Work in progress
> >> 
> >> For the particular case you gave, the FretBoards function works
> properly.
> >> 
> >> If the FretBoards function works properly to give the tablature you
> want,
> >> I'll adjust the tab-note-head engraver so it uses the same note-to-fret
> >> function that the FretBoards engraver uses.
> >> 
Almost all chords I tested were transposed properly (see caged.ly). Two chord cannot be played after transposing it (<c e g c g'> and <g c e g c>).

Two minor observations:
1) some of the chord names aren't centered above the fretboard diagrams. (Maybe this is a known issue. I didn't look for a workaround)
2) In the appendix B.2 Common chord modifiers (Notation Reference) there is a wrong chord modifier (and a wrong chord) for the minor/major seventh chord. It says 'maj7.5-' but it should be 'maj7.3-' 
 

> >> I think this is the proper thing to have happen, anyway.
> > Sounds good to me. By the way: Would you mind me adding (quite) a few
> fret
> > diagrams to predefined-fretboards.ly? If yes: is it possible to add
> several
> > diagrams of the same chord e.g. C major: A-shape (III. position),
> G-shape (V.
> > position), E-shape (VIII. position) and D-shape (X. position). I could
> also
> > add lots of inverted chords, eleventh, thirteenth and so on. Why are
> there two
> > separate fretboard files?
> > 
> 
> I would be fine to have you add diagrams to
> predefined-guitar-fretboards.ly.
> 
> We have two separate guitar fretboard files because we want to show in the
> documentation that if a chord is not included in the predefined
> fretboards,
> a chord shape will be automatically calculated.
> 
> It is possible to add several diagrams of the same chord by using
> different
> octaves.  However, the octaves should probably be two apart, e.g. c and
> c'',
> but not c'.  Because of this, I would recommend that we *not* have all of
> the possible shapes for each of the chords in the default file
> ly/predefined-guitar-fretboards.ly.  I think we'd fill up the "octave"
> name
> space.
> 
> If you can think of a good classification scheme, perhaps we could put
> these
> chords into separate .ly files, and the user could then include them as
> desired.
How about this one (see chord-classification-toc.pdf)? All the parts would go in separate .ly files. The chords with the question mark could also go in a different category…
> 
> If you're going to add different chords with the A shape, E shape, D
> shape,
> etc. you should probably do it with addChordShape and offset-fret to  use
> the known shapes.
> 
> > patrick
> > 
> > P.S.: I always get two identical e-mails from you (but not from the
> others). I
> > thought the mailing list prevents "double postings"?!
> 
> The gnu lists do, but the lilynet.net may not.
> 
> I'll try to remember to eliminate the reply to you directly, and just send
> it to tablatures.
> 
> Thanks,
> 
> Carl
> 
> 

-- 
Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3.5 -
sicherer, schneller und einfacher! http://portal.gmx.net/de/go/atbrowser

Attachment: chord-classification-toc.pdf
Description: Adobe PDF document

Attachment: caged.ly
Description: Binary data