Re: Odd output

[Michael Ellis]

Hi James,

To beam or not to beam should be about rhythmic grouping.  Stem direction tells which note heads go with which voice.  I think C is better for both vocalists and pianists. With or without beaming, it's clearer what's going on, especially with regard to which voices have the sharped or natural notes.  Slurs, AFAIK, come lower in the hierarchy of which marks have to make way for what.

 As to the second natural, if given I think it should be a parenthesized courtesy accidental.  At least that's my understanding of what they tried, and perhaps failed, to teach me in music school.  

Be that as it may, if I were composing for a particular instrument and ran into this situation I'd be inclined to consult some expert players and ask them which they find easier to read.

  
Cheers,
Mike

On Fri, Dec 17, 2010 at 6:09 PM, James Lowe <address@hidden> wrote:

Hello,

-----Original Message-----
From: lilypond-user-bounces+james.lowe=datacore.com@gnu.org on behalf of Michael Ellis
Sent: Fri 12/17/2010 21:50
To: Keith OHara
Cc: address@hidden

Subject: Re: Odd output

+1 for option C
Cheers,
Mike


On Fri, Dec 17, 2010 at 4:43 PM, Keith OHara <address@hidden> wrote:

> On Fri, 17 Dec 2010 04:09:10 -0800, Phil Holmes <address@hidden>
> wrote:
>
>>
>> The version that Chappell uses in the Mikado is attached.
>>
>>  Nice.
> It does break the usual rules about horizontal placement, and about when to
> cancel accidentals in another voice. But, with the beaming to clarify the
> timing and linking the voices, I did not notice any rules were broken
> --until I tried to produce it with Lilypond.
>
> It would be reasonable to ask Lilypond to produce 'A' below on her own, and
> even better if she would offset the notes as in 'B' (which she does do for
> chords in one voice).  Probably a human should decide when to bend the rules
> to produce 'C'.
>
> I think it helps to show the desired behavior in the tracker, and plan to
> put what is below in a comment to 1134, unless somebody either beats me to
> it or has second thoughts.
> -Keith
>
>
> \relative c' { \time 2/8
>  << s1*0^wrong
>    { fis8 g } \\ { f f } >>
>  << s1*0^A
>    { fis8 g } \\ { f! f } >>
>  << s1*0^B
>    { fis8 g } \\
>    { \once\override NoteColumn #'force-hshift = #1
>      f! f } >>
>  << s1*0^C
>    #(set-accidental-style 'voice)
>    { fis8[ g] } \\ {
>      s64 f!8*7/8[ f] } >>
> }

---

What about when extra staff notation is needed?

For example if you needed slurs or ties?

Wouldn't A be preferable here? That is having an accidental either side of each note is far more clumsy than two accidentals to the left of the note then the slur doesn't interfere.

Also what is the purpose in the case of A B or C of the second natural? Isn't that implied by standard notation where the note retains the 'sharp/flat' for the duration of the measure unless explicitly changed?

Here is a simplified example of what Keith did above to illustrate the point.

\relative c' {
<< s1*0

#(set-accidental-style 'voice)
{ fis8[ g] } \\ { s64 f!8*7/8[ f] }
>>

<< s1*0
#(set-accidental-style 'voice)
{ fis!8([ g] } \\ { s64 f!8*7/8~[ f] }
>><< s1*0
#(set-accidental-style 'voice)
{ fis!8[ g]) } \\ { s64 f!8*7/8[ f] }
>><< s1*0
#(set-accidental-style 'voice)
{ fis!8[ g] } \\ { s64 f!8*7/8)[ f] }
>>
}

I am not a vocal specialist but just using this one simplistic example of C seems erroneous. Isn't the idea of the notes printed at the same moment to show that they need to be sung at the same moment if you see what I mean? Yes I am sure that a vocalist can make their own mind up, but if that is the reasoning then it doesn't matter what we use then does it and you can provide instruction accordingly.

I don't think that the beaming clarifies anything at all personally.

Just my tuppence worth.

james