Figured Bass Continuation Lines Don't Work

• Oct 13, 2015 - 01:09

When entering figured bass, I use the underscore ( _ ) as directed, but it doesn't change into a continuation line when I space to the next note. It just remains an underline character, and the length is constant (doesn't change to the length of the group). Exiting the entry of figured bass doesn't change the underline characters either. I'm using MS 2.0.2, Mac OSX 10.8.5. See attached screen captures.

Also, how do you indicate chord changes on a single note? If you have a whole note with a 7 and then need to change to a 6-5 at the half-note mark.


Please post the specific score you are having problems with and precise step by step instructions to reproduce the problem.

See the Handbook under "Figured bass" for the shortcuts needfed to advance by a specific duration (eg, Ctrl+6 to advance by a half note).

In reply to by Marc Sabatella

Thank you for pointing out how to extend the groups, which solves part of the problem.

I could not get "Control-#" to work, however "Command-#" seems to work in the same way, so that's what I've done here.

If I press Command-G, a 6, then "_", then Command-3 to advance the group by 1/16 note, I actually have to press Command-3 twice to get the line to extend under the 1/16th note following the first. This does not make sense to me. See example 2 and the score, bars 2 and 3 of the first movement.

I was also able to apply this to example 1, bar 13 of the second movement. I cannot seem to get the continuation line to extend for the duration of the last note in the measure (and the 7), no matter how many times I attempt the extend the group or type additional underscores. It remains the same length. It seems the continuation lines do not end where the grey group lines do. It's rather inconsistent.

However, my main issue is the vertical positions of the continuation lines. They are too thick, and they appear below the numbers, like an underline. They don't look like the continuation lines that are pictured in the handbook. The proper style is to have them positioned where an em-dash would be, separated from the numbers with a thin space. I imagine I could correct this with a font editor, but I do think it would be a good idea to improve the appearance of these lines.


Attachment Size
Example1.png 22.24 KB
Example2.png 33.25 KB
Neuviéme_Concert.mscz 40.18 KB

In reply to by Soolip

FYI, pretty much any time you see Ctrl mentioned in documentation for cross-platform software, you need to mentally substitute Cmd, Similarly, when you see documentation mentin right click, on Mac that is usually Ctrl click.

As for the line extending, each time time you press a duration key, the cursor advances that far. This is separate from using underscores. But when using underscors, I believe they won't appear until advance far enough to need one. Perhaps I misunderstand, though. If you are still having problems, please give precise step by step instructions to reproduce the problem using your posted score,

As for vertical position, I don't understand. Whe I load your score, it looms like ypu describe it wanting to look, not like your picture shows. Perhaps you have some sort of font conflict?

In reply to by Marc Sabatella

I cleaned up the fonts on both my computers. Also, I opened up the file on a computer that has hardly any fonts installed, and I'm still getting what I had before — thick continuation lines that are positioned below the baseline of the numbers (see attached scree grab). Also, please note that continuation lines at the ends of measures and systems don't fill out the group completely, but end abruptly under the note.

I don't know what Ubuntu 14.04 is.

Attachment Size
Screen Shot 2015-10-15 at 8.09.55 AM.png 19.49 KB

Sorry for not having noticed this thread before. A number of points to note:

1) Multiple f.b. elements under the same bass note: to advance the f.b. cursor to the required point, use the [Ctrl]+<digit> keys as explained here . Once the proper 'mid-note' point reached, enter the f.b. element normally.

2) Continuation line yes/no: the underscore ('_') character after a digit only tells that a continuation line is needed; the actual line length is determined by the duration of the figured bass element; such duration is set in the same way as described in 1) above. The full cont. line might not be drawn completely and immediately while entering it, as the drawing algorithm needs some info depending on the next f.b. chord; usually, the line is drawn properly as soon the next f.b. element exists.

3) Extended continuation line: your "Example 1" above shows a continuation line which connects with the cont. line of the next chord. By default, the cont. line stops before reaching the next chord; to connect it, enter two (or more) underscores after the f.b. digit, as explained here . Of course, the duration of the f.b. element should be set to reach the next f.b. chord. This works only if the next chord also has a cont. line at that position with nothing in between.

4) Line position, thickness, etc: the 'font' has nothing to do with them, as cont. lines are drawn as graphic primitives (drawing lines); their thickness, vertical position, etc are hard coded and cannot be changed. They should not change at all from OS to OS, though; if they do, something else is involved which should not. The currently defined position (at digit base line) matches the position used in many modern scores I have used.

Agreed, in XVIII c. sources, different positions occur, also different from source to source, as well as non-horizontal lines and so on. I made the decision to uniform these parameters to one configuration commonly used in modern editions; it can be fine-tuned, though, if the need arises; to my eyes, the current result looks good once printed, but I agree that there is much of subjective in this.

Note: As said in other posts; keys like [Ctrl] have to be adjusted to the ones relevant for each OS.

In reply to by Miwarre

Thanks Miwarre for this explanation, it helps me to better understand

- A single underscore never works for me, I need to enter a double underscore to get any continuation line even in simple case (see measure one of attached test_figured_bass: single underscore and no continuation line). Snapshot captureFG.png.
- When there is no figured bass following the continuation line (implied 5), I get no continuation line at all in exercice_54 (issue 99861) or a line going "almost" under the note without figured bass (third measure of test_figured_bass, also in snapshot captureFG.png) which is of course incorrect.


Attachment Size
CaptureFG.PNG 13.31 KB
test_figured_bass.mscz 4.33 KB

In reply to by frfancha

1) Hmmm, well, in your example all the continuation lines are actually useless, as they span a single bass note. It would be similar to entering a melisma under a single note: it has to be expected that it may behave unexpectedly...

2) In your example, the bass notes are in voice 2, while the continuo figures are in voice 1 and match the pauses in voice 1.

In reply to by Miwarre

"continuation lines are actually useless as they span a single bass note":
I'm surprised by this remark, they are in the contrary very necessary: they mean that first half of the note has an harmony and second half another one:
E.g. 5- replaces 5 6 and 6- replaces 6 5

"In your example, the bass notes are in voice 2, while the continuo figures are in voice 1"
I will have access to Musescore this evening only and will test then according to your remark.

In reply to by frfancha

I am surprised by your surprise... <g>: in your example there are 5 bass notes (+ the last one) with one continuo figure each; one note one figure, one figure one note, nothing is "carried over" from one bass note to another, so there is no need for continuation lines.

(And, incidentally, your example has no '6' at all...)

In reply to by Miwarre

Again: 5 followed by a line under the half note C means a quarter e-g-c (5) above the C, followed by another quarter e-a-(e or a) : 5 followed by 6, and vice versa for 6_

Regarding the fact that numbers are attached to voice 1: indeed.
How shoudl it be done?
Selecting the first note of the voice 2 and press Ctrl G musescore switch to voice 1, if I'm first in note entry mode, ctrl G with first voice 2 selected also makes jump to voice 1.
What is the correct way to do do ctrl-G and stay in voice 2?

In reply to by frfancha

In fact needing to enter 6 (or5) followed by line under a single note is a little bit as the end of the original question of this topic:
"Also, how do you indicate chord changes on a single note? If you have a whole note with a 7 and then need to change to a 6-5 at the half-note mark"
Except that instead of 7 then 6-5 here I need 6 then _ (or 5 then _)

In reply to by frfancha

"5 followed by a line under the half note C means a quarter e-g-c (5) above the C, followed by another quarter e-a-(e or a) : 5 followed by 6": this is entirely new to me (as far as I know, if 3/5 then 3/6 is intended, usually "[nothing]   6" is written or occasionally "5   6", while continuation lines are only used to keep the same chord above a changing bass note) and might well be my ignorance. Are there historical sources for this interpretation?

In reply to by Miwarre

It is used in the Jean Doué Etude de l'harmonie, and I have read it also on Internet (which doesn't mean much, I know).
My harmony teacher uses the Koechlin as reference, I will check that in the Koechlin next lesson and will let you know.

ADDENDUM: looking again, the digit width set in the font does not seem to be observed while drawing; indeed the font data include more space between the digit and the cont. line.

As it was working once upon a time (at least when the handbook examples have been created) and that part of the code did not change, there is probably wrong with the font management at a lower level than f.b. management. I'll look into it.

In reply to by Nicolas

Well, not definitely.

However, I noticed that:

1) These cont. lines do not follow changes to page scale ("Layout | Page settings... | Scaling"). If the screen-shots posted above are made with different page scale, this may account for the difference in line thickness in comparison to character scale.

2) There is a difference in how the font is selected in the FiguredBassItem::layout() method for layout computations and in the FiguredBassItem::draw() method for actual drawing, which may result in differences in the respective position of digit and line at different scales. Even if I doubt this may account for the entire differences in the above screen-shots.

I have filed an issue for them: #84676: Figured Bass: continuation line not scaled to page scale and I have a fix almost ready.

3) There is the 'usual' problem with Qt rounding of font sizes / metrics which introduces a margin of unreliability in metrics data upon which characters and line are positioned. The fix above tries to use the way which, according to my tests, gives the smallest variation, but a margin is still there, as shown in these test screen-shots:


Presumably, the way to go is to drop Qt font entirely and use freetype for (some?) textual items too.

Do you still have an unanswered question? Please log in first to post your question.