[Top][All Lists]

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

Re: Is Anyone Working on a Better Tablature Algorithm?

From: christopher-heckman
Subject: Re: Is Anyone Working on a Better Tablature Algorithm?
Date: Thu, 10 Nov 2016 13:43:29 -0700 (MST)

David Kastrup wrote
> christopher-heckman <

> ccheckman@

> > writes:
>> In the past few years, a couple of algorithms have been released which
>> do a better job for tablature. Since Lilypond uses one of the most
>> naïve ones (not much different from the one that TuxGuitar uses), it
>> might be time for an update.
>> Is anyone working on this? I haven't been able to find any information
>> on current projects.
> Personally I think there should be a _translator_ doing the string/fret
> assignment and recording it into the music expression.
> Why?
> This translator can keep context and, for example, default to using the
> identical assignment for repeated chords, continued _notes_ and similar.

I strongly disagree with you on this. If you are playing the guitar,
especially if you are playing passages with lots of notes, you may not
always want to go to the same fingering for the chord. There are cases where
you would have to shift your hand position by 5 or 10 frets, play one chord,
and move back. This is not idea.

> It can also be employed at TabStaff level (rather than at TabVoice
> level) in order to create a workable assignment for polyphonic
> guitar/lute play.

The algorithm I have in mind does work for polyphony, so it can be used at
the staff level. (And that would be how I would design it: for the entire
piece.) It would accept a musical expression and then add the string (and
fingering) information to the notes.

> The output of the translator would obviously be also usable for
> fretboard diagrams and similar.

You're thinking of chords here, and I'm thinking of melodies/riffs. In that
case, fretboard diagrams really don't help.

> A "reverse" translator (quite simpler) could optionally translate
> string/fret to note pitch.
> The advantage of that would be that people could enter information in
> tablature (namely as string/fret pairs) and manage to create tablature
> as well as notes from it.  Currently you are restricted to entering
> stuff as notes.

Someone could work on that. It would be much easier than figuring out the
finger positions.

> Some combination could even be used for letting people enter music in
> tablature.
> So with a bit of retooling, one end up with a more versatile set of
> modular tools than what we have now.

Which begs the question that I started off with: Is anyone working on this?
We shouldn't have half a dozen people working on the same improvement
independently; you will get half a dozen different pieces of code, all
inconsistent. (And if one of them works for Microsoft, you may get eight or

--- Christopher Heckman

View this message in context:
Sent from the Dev mailing list archive at

reply via email to

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