Re: GDP: NR 1.1 Pitches vastly improved, more comments sought

[Graham Percival]

On Mon, 21 Jan 2008 11:24:22 +0100
Mats Bengtsson <address@hidden> wrote:

> Graham Percival wrote:
> > Some sections have been completely rewritten (particularly Octave
> > check).  Please read the new Pitches section and send comments.
> >   
> - The text in "Octave corrections and checks" is contradictory.

Whoops!  My initial impression of these commands was completely
incorrect; after creating the example shown on the page, I
realized my mistake and fixed them... but apparently not
everywhere.

> As far as I can see, the true difference between the two
> commands is that the
>   octave correction is specified as a attribute to the note itself in 
> the input, whereas
>   the octave check is a separate "command" to be inserted between
> notes in the input.
>   Both have the same effect of correcting the octave (not the pitch, 
> right?) if it's wrong.

Not true; with = the d's octave is changed; with \octave the d's
octave is not changed.

Now, whether this is _deliberate_ or not is a completely separate
question.  :)   To be honest, I couldn't decypher the old docs on
this, so I just experimented.  (and I /remember/ being completely
lost when I wrote the initial docs three years ago...)


> - In "Transpose", I wonder if the second example is correct or if it 
> should have
>   an es major key signature (in the printed output, i.e. a \key c
> \major in the
>   input file), according to normal notation conventions.

Good catch!  I'm sure that it should; I know that conductor's
scores often have different key signatures for the weird (ie
transposing ;)  instruments.

>   Technically, this second example illustrates a "feature", namely
> that if you don't
>   explicitly specify a key signature, then the transposed music won't 
> have any
>   key signature either, in contrast to what happens if you explicitly 
> specify a
>   key signature.

Noted, thanks.

> - The transposing-pitches-with-minimum-accidentals-smart-transpose.ly 
> LSR example,
>   included in "Transpose" didn't work, since the internal
> representation for
>   alterations has changed between 2.10 and 2.11. I have just tried to 
> submit a
>   2.11 version to LSR, but since it doesn't compile in 2.10, I'm not 
> sure it
>   was accepted in the LSR database. I'll send it by email otherwise.

Please send it to me.  (I've complained to Valentin about this,
but if you already have the solution, great)


> - Instrument transpositions, first sentence: I'm not sure it's the
> "key" of the transposing
>   instrument that should be specified, but I don't know the proper 
> terminology.

I'm not totally certain, but I think that "key" is the correct
term.  Anybody?

> - I have submitted an even more simplified version of the LSR example
>   Ambiti multiple voices to LSR. The setting of X-offset didn't 
> influence the
>   result at all, as far as I could see.

Thanks!  Valentin, could you make sure it has the right tags, and
send me the filename?

Cheers,
- Graham