Re: Figured Bass question: Why is <5 4> upside down?

[Laura Conrad]

>>>>> "Trent" == Trent Johnston <address@hidden> writes:

    Trent> I don't think these are bugs as such. It's got to do
    Trent> aligning the extender line. If you place a spacer (s8 etc)
    Trent> between the figures the figures are printed the right way.


    Trent> is sent a reply to the Lilypond user list a while ago with
    Trent> a solution to this problem.

    Trent> http://lists.gnu.org/archive/html/lilypond-user/2006-07/msg00052.html

This message says:

    Trent> The figured bass figures are aligned so if you have a 6 4
    Trent> then a 4 3 the fours get aligned together so the 4 3
    Trent> becomes 3 4. To get around this put a spacer between the
    Trent> figures. For example if they are <6 4>8 <4 3> to get rid of
    Trent> the alignment use instead <6 4>16 s <4 3>8.

    Trent> This is not a bug but a feature for extender lines.

This seems like an undesirable feature to me.  It puts a lot of burden
on transcribers of figures to notice whether figures are repeated,
when most figured bass that I know anything about doesn't use extender
lines at all.  And transcribing figures with no support for MIDI
review is difficult enough without the figures being reordered on
you.  

In any case, even if there were an extender line, you would still want the
numbers to be in the correct order; you would just want the vertical
spacing to be different.  So you would still want <4 3> and not <3
4>.   If there were an extender, you would want

      6

      4 ____

            3

But if you weren't trying to extend, you would rather have

    6     4

    4     3

So that the figures would be closer to the notes.  

So I still think it's a bug.  The fact that it's a bug with a
workaround makes it less critical, but it's still not what anyone
would expect to see.   

Is there any way the behavior could be removed, at least for cases
when useBassFigureExtenders is false?

If the bug can't be fixed, I think the workaround should be
better documented.  Here's the current explanation;

    2.20> When using continuation lines, common figures are always put
    2.20> in the same vertical position. When this is unwanted, you
    2.20> can insert a rest with r. The rest will clear any previous
    2.20> alignment. For example, you can write

    2.20>   <4 6>8 r8

    2.20> instead of

    2.20>   <4 6>4

I think this isn't quite correct, since the problem currently occurs
whether you're using continuation lines or not.

I propose that it be changed to:

    new> When two immediately adjacent columns of figures share a common
    new> number, Lily sometimes gets confused about what order to set
    new> the second column in.  If this happens to you, you can remove the
    new> confusion by inserting a rest or spacer between the figures.

    new> For example, if you have

    new>     <7 5>4 <5 4>

    new> and lily is putting the 4 above the 5 instead of below, you
    new> can work around this problem by instead writing:

    new>        <7 5>8 s8 <5 4>4

-- 
Laura (mailto:address@hidden , http://www.laymusic.org/ )
(617) 661-8097  fax: (501) 641-5011
233 Broadway, Cambridge, MA 02139