Paired hairpins

• Aug 14, 2019 - 17:08

Not sure if this is a bug report, a feature request or a topic for general discussion.

Hairpins can be anchored to specific start and end notes, but what if you want paired hairpins on a single note. I'm trying to put crescendo decrescendo pins on a measure containing only a whole note. Ever since the last update I've had trouble adjusting the length of hairpins. I double click them and grab an end handle but when I stretch or shrink them they jump back to their original length.

It would be a nice feature to add paired hairpins to the lines palette (i.e. something looking like <> and ><) in order to anchor that to a whole note.


I would second the idea of dedicated "paired hairpin" lines. Having to attach/anchor every line to a note or rest has always (even in Sibelius, if memory serves) made this kind of thing a bit onerous in relation to how common it is. Not that it's bad or unwelcome advice, but any workaround that contains "enter rests into an unused voice" indicates a less-than-ideal state of affairs.

In reply to by kacattac

This is what we have, I didn't write the code. I have seen some discussion of allowing for hairpins and lines to have end points that are not on notes, but it hasn't happened to this point and I don't know when it might if ever.

If you don't like the workaround you can either write the code to fix it or drag the end points and live with the consequences that include they will not play back properly and possibly that they will move around when you save and load your score.

In reply to by mike320

"... and possibly that they will move around when you save and load your score"
That's a very important point. By attaching the end point or start point of the hairpin to a mid-measure rest, we can get a much more reliable placement of the hairpin - it doesn't end up "wandering".

I think this is a good suggestion, and adding paired hairpins as a new sort of line should be a trivial job. It's depressing that at least half of respondents to this post do not seem to have read it carefully, and are talking about all sorts of other things.

An ordinary hairpin has three parameters: start point, end point, and height. A paired hairpin could be defined by exactly the same parameters -- the only other variable could be the relative length of the gap in the middle. I would guess that this could be fixed at (say) 5% of the length, and this would satisfy almost all requirements, without needing any extra functionality other than the new line shape.

In reply to by Imaginatorium

I have read what you said, I simply explained how to accomplish what you want in MuseScore. Your solution does not allow for putting a dynamic in the middle of the paired hairpins, which is not at all uncommon. Also, paired hairpins do not always have a break in the exact center, one may be a measure long and the other 1/2 a measure long, this issue has not been addressed all A solution that allowed for arbitrary length hairpins would eliminate the need for using hidden rests.

In reply to by mike320

But for the case where the first hairpin is a measure long and the second is 1/2 a measure, there would be separate notes to anchor two hairpins. The difficult situation is when you want an up and a down hairpin pair on the same note.

I have used the "volume swell" from the articulations palette sometimes. Volume Swell.mscz It's not pretty, the length is not adjustable, it doesn't affect playback, and it's placement needs tweaking to get it centred under the note, but it sort of does the job.

In reply to by SteveBlower

There is only a note for the endpoint of the hairpins line up. Not if there are two whole notes tied together and the crescendo starts on beat 3, you then only have an end point on the end of the last note.

Volume swells are very unacceptable for use as hairpins, as you have discovered.

In reply to by mike320

If cresc. and dim. hairpins follow each other on a single note then it seems that the composer/arranger is leaving it to the performer when the transition from cresc. to dim. occurs. If the composer specifically wants it to happen at a particular point, then that must be specified by splitting the single note so that there is a note that actually starts when the transition is wanted.

The case that you describe is similar if I have not misunderstood what you are saying. If the cresc. is to start specifically on beat 3 then that must be specified by having a note that starts on beat 3 and the start of the cresc. hairpin can be anchored to it. In this case, perhaps with a dotted 1/2 note tied to a 1/4 note so that the start of the hairpin can be anchored to the 1/4.

In reply to by SteveBlower

Composers do not break down notes and tie them to accommodate dynamics. They write the entire note to be played and then stick the hairpins in so the look like they're on about a certain beat. The conductor can usually tell by looking at the score where the hairpin starts based upon where it is on other instruments that do have notes on that beat. The musician knows from looking at the hairpin that he needs to watch the conductor (or ask) to get the cue for the dynamic change. That's why if you want it to look reasonable in MuseScore you have to use invisible rests to anchor the endpoints.

In reply to by mike320

Gould, Chapter 4, Dynamics, Horizontal placing (page 104 in my copy):

"A hairpin will always be a very precise length: this is its great advantage. Therefore it must be carefully placed or its significance will change.

Do not place a hairpin before a note is started nor after a note is finished. A hairpin should start at the first relevant notehead (not accidental) and end with the following notehead or at the first rest thereafter. Good practice is to start the hairpin on the left-hand edge of the note and to finish it on the right-hand edge of a note.

(If a dynamic symbol is present, the hairpin starts later and finishes earlier so that the dynamic centres on the notehead or chord.)

When precision is important, subdivide note -values so that dynamics can be placed precisely"

(my emphasis)

There follows the attached example which demonstrates what I said in me earlier post. Hairpin example.jpg

In reply to by SteveBlower

Gould sometimes fantasizes about realities that don't exist. The "precise" example you posted should get a publisher fired for putting notes in that are unnecessary and confusing to musicians. Are these supposed to be ties or phrasing marks (that look exactly like slurs and ties)? The "vague" example is what a musician expects to see when they are to play a whole note. That's also why there is a conductor to put all of the musicians on the same beat.

If it is ever made so lines (including hairpins) can start and end on any beat there will be no need for mirrored hairpins. This would be a far better solution than special casing one line type and ignoring the rest.

In reply to by mike320

I don't think she is suggesting that publishers convert music that doesn't already look like this into music that does. But it's good advice to a composer who needs to be that precise - if for example the entire effect of the measure is based on that sort of precision. Most composers don't write that way of course, but it remains good advice if you do need that sort of precision. This sort of precision is that uncommon in "modern" scores.

Returning to the initial suggestion: I add my support for the idea of paired cresc.+dim. and dim.+cresc. hairpins that can be entered with a single anchor point but with adjustable length and adjustable central gap. The ability to insert a dynamic in the centre would be a good enhancement, but probably would add significantly to the difficulty of implementation.

Perhaps paired hairpins without the central dynamic facility could be implemented as a first step and then the central dynamic added later.

In reply to by SteveBlower

Yes, I think it would be a good idea to add the simplest form, with no variability over and above that of an ordinary hairpin. I have already said this above, but my comment got hijacked into irrelevancy. Anything more complicated can (or possibly should) be done with separate hairpins; the point is that this would speed up the job of adding the fairly common "get louder in the middle" notation, without adding any implementational complexity.

I began a newer thread on this topic and was then alerted to this older thread on the same subject. So it looks like I'm not alone in desiring a simpler solution for this common function.

I have one idea that will take longer to describe than its relative simplicity would suggest. It would provide a series of options, the first being a paired hairpin that is centered within a measure. This would be the default version. Variations would consist of other hairpin pairs that would allow the user to define the start and end points of both segments of the hairpin, based on beats (or implied beats if, for example, the measure consisted of just a whole note). Both the default configuration and the variations could be assigned shortcut keys that would allow the insertion of exactly what you wanted with two keystrokes. Certain details would need to be worked out, but more complex things have been done. Just a thought.

In reply to by remmet

I don't think that defining the "join point" in terms of beats is necessary. As I argued when this thread was first live, if someone wants to specify the exact time where the cresc. stops and the dim. starts, then it is necessary to actually have note divisions to attach the endpoints to, and, more importantly, to let the performer see those notes. This much we can already do. What is needed is the option to have a swell but not specify exactly where the join comes - i.e. cresc. and dim. over a single note. I think a good solution would be to have handle which defaults to mid point but that can be moved horizontally to any arbitrary point, not tied to beats (or 1/2 beats or 1/3 or 1/4 beats etc. - just think of the complication required to implement those!) to indicate slow cresc, fast dim. or vice versa as desired. If a dynamic could also be attached to that handle, that would be even more splendid.

Hello all. This is the only conversation on the topic of paired hairpins I can find. Please let me know if I'm completely missing an already existing solution. I am constantly having to manually adjust hairpin veritcal positioning so that adjacent cresc - dim hairpin pairs are vertically aligned so they look nice (to be clear, for now I am just talking about wanting automatic vertical alignment when a dim hairpin starts on the note immediately after the end of a hairpin cresc). Seems like this could work the same way that automatic alignment for adjacent dynamics & hairpins already does. Please let me know if there is already a solution or consider this a request for a new feature. Thanks.

In reply to by msfp

Hello and thank you for this suggestion. Is this plugin still "adding a lot of small undoable steps that almost renders the undo function pointless."?

It seems like a great feauture to have, but useless if you loose probably one of the most useful action live the undo function.


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