Incomplete tuplet brackets in irrational meter

Hello - I know this issue has been raised before but there were multiple issues in the relevant threads.

Currently, Dorico shows a measure of, for example, 2/6 as having 2 crotchets, without any tuplet brackets. This is in a very pure sense ‘correct’. However, it doesn’t conform to anyone’s actual practice as far as I know, due to players’ understandable wish to play what they see - i.e. if they see a crotchet, they play it. In all working scores of this type (Ades etc) some kind of indication is given that the crotchets are faster. The most frequently encountered method, and the one that is most easily understood by classically-trained players, is the incomplete tuplet bracket:
incomplete tuplet.png
I understand that there is no native support for this on Dorico (despite the admirable ease of input of irrational meters), and I am interested to see if anyone has found a good workaround.

One way would be for Engrave to permit the hooks on a tuplet bracket to be adjusted individually - at the moment both sides may be hidden but not just one. Even this approach would need some workarounds as Dorico interprets the added tuplet bracket to be in effect a nested tuplet (within the 2/6 time signature), and so the extra rests need to be made transparent.

Another idea is to input the incomplete tuplet bracket graphically. This of course will have its own problems - can only be done when every other aspect of page layout has been set, for example. And I can find the tuplet number glyphs in the SMuFl but not the bracket elements. Does anyone know where these might be found?

Lastly I thought of graphically adding a little white box to erase the tuplet hook. Very time-consuming for a large score…

Or there’s always a printer, Tipp-Ex, and a photocopier… but it would be nice to find a more elegant solution!

I’d be very grateful if anyone could help with this.

When we eventually come to work on this area some more, including handling the appropriate metric modulations to define the relationship between the irrational and surrounding rational meters, we may be able to handle the drawing of incomplete tuplets as part of that work, though it’s not yet scheduled.

There aren’t any glyphs in Bravura for drawing tuplet brackets that would be any use: Dorico draws them as primitives. You can’t adjust the hook length of each end of the tuplet bracket independently, so unfortunately you can’t use a real tuplet.

I don’t think there will be a good way of doing this that isn’t pretty labour-intensive, I’m afraid.

Vector drawing apps, like Adobe Illustrator, (or Affinity Designer, or Inkscape…) can open PDFs, edit the graphical elements, add text, lines and curves and then re-save the PDFs.
Many people use these to overcome limitations of notation apps or to produce more fluid ‘drawing’ styles of notation.

Note that Photoshop and other photo editors are not the same thing at all, and not suitable.

If the user has stipulated “irrational” time sig denominators, the brackets are not really tuplets anyway, just a visual aid (bracket/line) for the performer; so they don’t really have t be functional tuplets if the time signature is functional.

Thank you, Daniel, Ben and Derrek.

Derrek is quite right in saying that these are not true tuplets, but only graphic features. However, they are indispensable for the player, and I’m seeing more and more scores with this type of rhythm (often not to write very complex rhythms but just to avoid the clumsiness of having a metric modulation for, say, a single 1/4 measure). So I guess a certain number of composers have found a solution, probably in a vector app as Ben suggests.

Daniel - would it be very onerous in the future to allow Engrave to adjust hook lengths individually? And/or to be able to draw primitive lines in the Playing techniques editor?

If it’s any help, I guess that most users of this feature would appreciate having the ability to draw these graphically first, and are not too bothered about sorting the metric modulations for playback.

Thanks once again to the Dorico team and user community for this helpful forum.

This was my thinking. If the time signature handles the actual playback, then the brackets can merely be graphic items, which would make them easier to implement unless I am mistaken.

Dear Dorico Team!
Incomplete tuplets is something I need to use often and would very much welcome their addition to Dorico soon.


Welcome to the forum, @mikesvoboda. Thanks for letting us know that this would also be a feature you would find useful.


Since this post was originally written, Dorico added support for drawing arbitrary lines. You can create this without much effort, though it may require some tweaking to the line body, hooks, and font in order to match the desired tuplet style:


Could you please indicate how to notate 2 “normal” crotchets in a 5/6-measure? According to Dorico’s current handling I would have to write 2 dotted crotchets, which is somehow absurd.

Do you mean (A) you intend to fill the measure with two crotchets, or (B) you enter two crotchets, but they play for the wrong length of time?

(A): I believe Dorico interprets these meters differently than what you are expecting. Denominators round to the next largest note value as if it were a tuplet of the semibreve. This means n/6 rounds to 1/4 (crotchets), and n/12 would round to 1/8 (quavers).

5/6 typically means “a measure of 5 beats, each of which is a sixth of the semibreve.” That way the player might see a series of standard triplet crotchets in 4/4 (which are each 1/6 of the semibreve), and when suddenly they see 5/6 meter, they can just play five of them without having to perform a metric modulation in their head.

To have two crochets in 5/6 seems erroneous, in the same way that filling a 4/4 measure with only two crotchets is. Nevertheless, if you want to do this, you can create a hidden tuplet:

  1. Round the denominator to the next largest note value and select it (in this case, crotchet)
  2. Open the tuplet popover.
  3. Enter x:y, where x is the target number of notes (2) and y is the numerator (5).
  4. Select the tuplet and hide the bracket and number in the properties panel.

This creates a tuplet that’s “2 crotchets in the span of 5”.

(B): Dorico does not perform the required metric modulation for you. You must add a hidden tempo change at the start and at the end of the measure.

  1. Multiply the current tempo by x/y, where x is the new denominator (6) and y is the new denominator rounded to the next largest note value (4).
  2. Add a tempo change at the beginning of the n/6 bar.
  3. Add another tempo change to reset at the beginning of the bar that follows the n/6 meter.

For example, when 4/4 at q=120 is followed by 5/6, you need to add a hidden q=180 tempo change at the beginning of the 5/6 bar, and add a hidden q=120 tempo change when going back to 4/4.

I am sorry to have been unclear. In the meantime I managed to notate what I had in mind (see attachment). However playback is wrong (no problem for me).
I might have had the wrong impression that Dorico would consider crotchets in a 5/6-measure as triplets which is apparently not the case. Therefore my question referred to the “duole”-rhythm in the left hand. I think my notation is the most practical one - despite the tautology of the 6-denominator and the triplets.
Thank you for your help!

Hi there

Your 5/6 bar looks pretty good to me, except that the downstem minim B-natural needs a partial triplet sign as well. (And you could adjust the brackets a bit to make them mimic the look and placement of true triplets, too.)

I don’t think there is anything tautological about requiring the triplet signs in the 5/6 bar. Time signatures tell us nothing about notes in their bars: they tell us about a bar’s duration and in most cases something about the metrical organisation, which is different.

My editing rule for these is that any tuplet signs that would be required for the music if it were in a non-irrational time signature are also required in bars with irrational time signatures.

Some of the tuplets will be partial ones, because otherwise the irrational time signature would not be necessary.



This might be an even better solution.!

1 Like

Hi again

The new time signature makes things worse!

There are three reasons. First, it’s an innovation where one isn’t needed, because there is a (new-ish) convention already established (6 as denominator). Second, use of note values as a denominators in time signatures should be consistent throughout a score, so that the musicians will be used to dealing with them. And third, the triplet number hanging below the staff is easily missed or misinterpreted as meaning something else.

Sorry about that!

1 Like

OK, thank you for your comment! Although I want to note that 5/6 is not that established yet (only among freaks :wink: ) - I always have to explain to my students what it signifies, nobody is familiar with it. I thought my solution was intuitive. Of course I should improve the layout (I am still a bloody beginner in Dorico :frowning: )

Yes. In terms of the evolution of music notation, it is very new. But it’s been around in published music for quite a few years (decades now) and it is not confined to use by one composer, publishing house or region. So I think that counts as approaching an accepted notational standard.

The sector of music within which it is found useful is very narrow, so it’s no surprise that most students would not have met it. And never need it.


Thank you very much for this, very helpful! With my best wishes from Basel, Switzerland, Mike Svoboda

Hello Mr. Hughes,

I am writing you again in this context because I noticed a mistake in the German manual:

Eine Taktart ohne Zweierpotenz ist z. B. 5/6, womit fünf Sextolen angezeigt werden, die zusammen die Dauer einer ganzen Note haben.

5 sextuplets are less than a whole note! Otherwise they would be quintuplets! – As we can see this suject is still a source of confusion!

Best regards!

Martin Lichtfuss

Hi- It’s true that new scores that use denominators like 3, 5, 6, 10, 12 etc usually—but not always—come with an explanatory note (as of 2020) but the convention itself is at least as old as Henry Cowell’s New Musical Resources back in 1930! More recently it’s used by composers including Claus-Steffen Mahnkopf and others since (at least) the 1980s.


As you say the denominator convention should be consistent throughout. However, the ‘note value as denominator’ approach has its merits—Boulez used it on occasion, and plenty of composers particularly from Eastern European traditions including Kurtag and Eötvös, including the little fake triplet (although it’s invariably above the stave so it doesn’t get mixed up, as you say). It looks a bit less ‘mathematical’ when read at speed.

Anyway, it would be great if Dorico had full support for all these options, although I’m aware it’s a niche request!