Hairpins filling measures
How do I format hairpin entry so they don't automatically extend past rests? I would think that I select a note, and the hairpin notates visually only for the duration of that note (for example, a dotted half note) and not including the rest after it unless I also select that rest when I add the hairpin. But Musescore doesn't seem to do that by default.
Also: how do I search the forums to see if this question has already been asked? I see two search options here, neither of which is to search posts: searching for sheet music and searching for the online documentation, only.
Comments
see https://musescore.org/en/handbook/3/hairpins#adjust-length
In reply to see https://musescore.org/en… by Jojo-Schmitz
No, I found that. I don't want to have to go and manually drag the hairpin length every time I add one where a rest follows and I don't want it to fill the measure.
In reply to No, I found that. I don't… by tromboneandrew
Use the keyboard shortcuts "<" and ">". Then the hairpin applies only to the note(s) you have selected. If you apply it by clicking on the pallet item it extends to the end of the measure.
I don't understand why there is a difference between applying a hairpin by clicking on a pallet item and by using the shortcut. Sometimes one method causes problems while the other doesn't such as #297444: Hairpin attached to note other than in voice 1 in score appears on wrong part and #319193: Undo will not remove some improperly added slurs for example.
In reply to Use the keyboard shortcuts "… by SteveBlower
Ah, thanks. I agree, that doesn't really make sense.
In reply to No, I found that. I don't… by tromboneandrew
Then enter them correctly in the first place, by not selecting a whole measure
In reply to Then enter them correctly in… by Jojo-Schmitz
Or select the dotted note. Then Ctrl + select the rest. Then add the hairpin. Hairpin will stop at the rest. This is in the manual.
In reply to Then enter them correctly in… by Jojo-Schmitz
But as I said, not selecting a whole measure only works if you use the shortcut. If you click on the hairpin in the lines pallet it applies to the end of the measure.
Here is the difference in a picture.
![Hairpin.png]( "Hairpin.png")
In reply to But as I said, not selecting… by SteveBlower
Range select the note and use the palette should also work.
In reply to Range select the note and… by jeetee
Yes, but why should not range-selecting a note before the operation end up with a different result when using the shortcut rather than the pallet? I would expect there to be some "add hairpin" code that gets triggered by either the shortcut or by the click on the pallet, but it seems there must be two different codes, one activated by the shortcut and a different one activated by the pallet click. This seems to be wasteful, make code maintainance more difficult than it need be and to cause user confusion.
If you follow those two links I posted you will see other cases where an apparent duplication of code results not just in different but arguably equally valid results as with hairpins but in one action triggering correct behaviour and another action triggering incorrect behaviour. And it is not consistent that the keyboard shortcut gets it right and the pallet gets it wrong.
In reply to Yes, but why should not… by SteveBlower
There is no argument on that respect, I was merely highlighting a 3rd scenario missed in the current comparison.
In reply to There is no argument on that… by jeetee
But range select does not work in the case of a single note, as far as I can tell.
The Ctrl + select works better for me because I'm a mouser. To use the keyboard shortcut > requires me to take my hand off the mouse.
In reply to But range select does not… by bobjp
It does: select note, Shift+right, Shift+left.
Or just Shift+select the single note
In reply to It does: select note, Shift… by Jojo-Schmitz
Yes. There it is. Shift+select the note indeed works.
As far as a hairpin on a single note goes (which I almost never use):
I agree that for most folks, the shortcut is probably faster. But not for me. I use a trackball mouse. So my right hand is not moving around at all.
So to use the shortcut, I have to:
1.Select the note.
2. Move my right hand to the shortcut.
3. Left hand (shift) and right hand (shortcut)
4. Move right hand back to the mouse.
If I use the mousepad, my right hand is at least closer. But I dislike the mousepad.
Or I can:
1. Left hand (Shift) and right hand (select note)
2. Select line from palette.
It just depends on the situation or what you are used to. For me it is more consistent to use the palette.
In reply to But as I said, not selecting… by SteveBlower
I would claim not selecting a whole range almost never "works", or always does, depending on your view. That is, if you don't select a range but instead only a single note, there is no possible way for MuseScore to know how long you want the hairpin to be. maybe you want just one note, maybe a whole measure, maybe three and half measures - how could MuseScore possibly guess? So it randomly picks something. It always picks something, so that's the sense it always works. but it will only rarely happen to be exactly what you wanted, so that's the sense in which it almost never works.
Bottom line: if you care about the range your hairpin applies to, select it first. Same for all other lines. otherwise, MuseScore has to guess, and it will of course be wrong most of the time, so you'll have to adjust.
In reply to I would claim not selecting… by Marc Sabatella
The difficulty here is that if you want a single note as a range, it is far easier to just click the note and apply the symbol (which is what the shortcut already does) than it is to range-select the single chord.
I get that this makes less sense for some line types; but in this case, the shortcut action already does the "expected" thing in all scenarios already. Performing the seemingly identical action by clicking the palette element does not.
In reply to The difficulty here is that… by jeetee
The question to me is, how often is a single note actual something you want to put a hairpin, or other line, over? I would guess that answer is, not very often at all, that's the whole reason these things are designed as lines that can extend over multiple lines. Probably the vast majority of hairpins in published music cover multiple notes, and that's probably true of pedal, ottava, and other lines as well.
So to me, the full measure is more likely to be the right guess than the single note is, but still, it's just a guess, and any guess is going to be wrong, a lot. I have nothing against making them consistent, but if comes to a vote, my vote is to always guess the full measure.
In reply to The question to me is, how… by Marc Sabatella
In terms of design, I think "how often is a single note actual something you want to put a hairpin" is the wrong question. The question should be "how can the interface be as consistent and as simple as possible, so users can know exactly how to get what they want, all the time."
In reply to In terms of design, I think … by tromboneandrew
Well, that much is already the case: simply make the exact range selection you want, the hairpin will then cover exactly that, consistently, every time. The only question is, what happens if the user for whatever reason fails to to provide the necessary information and the program has to guess instead. I agree it would be nice to guess more consistently. But then, I would definitely like to see it consistently guess in a way that is the most likely to be correct of any of the possible guess it could consistently make. And here, again, I think it's measure. If we have to guess consistently, that has the greatest likely of being right. Maybe as much as 30% of the time it would be right. Guessing single note would probably be right only 10% of the time. And then there's the other 60% of the time where neither of these are right. Completely made up numbers of course, but probably not terribly far off, either.
In reply to Well, that much is already… by Marc Sabatella
My point is that the interface is inconsistent. The result depends on whether you use a keyboard shortcut or click the pallet.
In reply to Well, that much is already… by Marc Sabatella
Never have to "guess". That's bad software interface design.
In reply to Never have to "guess". That… by tromboneandrew
Accept that depending on how you use it, the interface is completely consistent. Just like most everything else, there is more than one way to do things. If you use the CTRL+select range, it dose exactly what you want every time. I didn't even know that was a thing until yesterday. We are all set in the ways we do things. I get it. We may find CTRL+range to be inconvenient, but it is a single way to add hairpins. I don't have to remember different methods depending on whether I want one note, on measure, or several measures. Each of which can be done a different way. Maybe you youngsters can remember all those, but at my age I'm lucky to remember pants.
In reply to Never have to "guess". That… by tromboneandrew
If you don't specify the exact range, we have to guess. There's no way to know. That's my point. The only way to not guess would be to do nothing - to refuse to place the hairpin at all if you don't make the full selection. I don't think anyone would prefer that. You absolutely want the hairpin placed, but since you didn't say how long you want the line we to be, we need to gues,s there is no way around that.
And yes, it should guess consistently, I agree and have said so several times. I am suggesting we make the guess that is likey to be correct the most often - maybe 305% of the time if we're lucky. And that's guessing "full measure".
In reply to If you don't specify the… by Marc Sabatella
I don't understand why you say someone doesn't specify an exact range. This is made specific exactly by what the user literally selects - whether it's one note, three notes, two notes and a rest, one measure, etc.
In reply to I don't understand why you… by tromboneandrew
A single note is not a range. A range has a specific definition in MuseScore - it's the thing you do that results in a blue rectangle displaying around the range. The range specifies a start and end point, and by definition includes all points between. A single note isn't a range, it's a list (of one). It's precisely the same distinction that exists between a line and a point in geometry.
So, if you select a single note (point), we can assume you meant that to be the start point for the line, but in order to actually draw the line, we need to know what other point you wish to use as the end point. Since you haven't told us, we have no choice but to guess. And as I've been trying to explain, it really does help with usability if you make the guess most likely to reflect reality.
I got curious and started thumbing through my copy of the Norton Anthology of Western Music looking for hairpins to put numbers on this. I deliberately skipped piano music because there will obviously be no single note hairpins there, thus trying to bias the numbers in favor of the possibility that single note hairpins might turn out to be more common than hairpins that happened to go to the end of the measure.
But no, turns out I was eerily close to correct with my initial wild guess, at least with respect to the "end of measure" number. Turns out I laughably overestimated the number of cases where favoring single note would produce the best results. I looked at 80 hairpins scattered across a number of pieces across different centuries before encountering my very first single-note crescendo. And that single note crescendo happened to be a note that filled the measure (dotted half in 3/4 time), meaning either algorithm would succeed as often,. I had to look at 100 more hairpins before I found my first example of a singe note hairpin that did not happen to extend exactly to the end of the measure.
In the end, I did the math, and my guess of 30% for number of cases the end-of-measure algorithm guesses correctly turns out to have been a little low - it was 38%. My guess of 10% for the number of cases where a single-note alogirthm would produce the right answer was a bit high - it's not 10% but around 5%. However, of those, the vast majority turned out to be cases where it was indeed to end of measure, meaning either algorithm would produce the right answer. The number of cases where the single-note algorithm produces the right answer but the end-of-measure algorithm does not was well under 1%.
So, I am now more confident that in cases where MuseScore is forced to guess because the user did not specify the range, guessing end-of-measure is absolutely without any possibility of doubt the statistically best answer, by far. Achieving 38% success in a guess here is better than I had even hoped for, and quite obviously no other algorithm would do better - the remaining 60% or so were "none of the above" in terms of length. So were three or four notes within a measure, some extended multiple notes, etc.
In reply to A single note is not a range… by Marc Sabatella
A single note selected means a "range" of that single note, no matter what "range" means for a multi-selection in Musescore. If someone wants a hairpin to fill the measure, they select the measure. The cases in historical notation do not have much bearing on the actual mechanism to notate in software for the mechanism to be applied consistently and users can always know what the software will do. Good software interface design does not involve "guessing" at all. The opposite, in fact: good design trains the user to input what they want in a specific, repeatable way, so there's no ambiguity. I don't understand what is so unusual about this concept.
Here's a similar "guessing" scenario that is even more obviously wrong: enter a triad, starting with C-E. If the software sometimes "guesses" the 3rd note is always G and just enters it before giving the user any other options, because of historical practice, this would be roundly trashed by everyone who tried to use such software.
In reply to A single note selected means… by tromboneandrew
Maybe in some other program a single selected note would mean a range, but MuseScore doesn't work that way. Nor should it, in my opinion, for exactly the sort of reason I just explained with my analysis. It yields the wrong answer way way way way too often.
You say the software shouldn't be guessing, and then go right on in the next breath to say, if a user clicks a single note, it should interpret that as meaning a range selection encompassing that note and that note only,
That is a guess just as surely as any other guess. And paradoxically, in the very next breath, you also point out exactly why this doesn't even work: there is no such thing as a range selection containing one note only in the case of a single note selected within a chord. That, again, is not what a range selection is. There is absolutely no way to have it both ways - to treat the single note as if it were a range selection, and to allow a single note to be selected within a chord (for the purpose of, say, changing its pitch). There is a reason list and range selections are different things in MuseScore, and this example illustrates it well.
So anyhow, yes, it would be possible to redesign the program to work more the way you personally prefer, to guess the way you want it to - that single note selection really is intended to mean a range of that note only. The end result would be, now only 5% of cases would actually work as desired instead of 38%. I don't see the payoff in that. Usability matters. Getting the right results by default as often as possible is a good goal.
In reply to Maybe in some other program… by Marc Sabatella
I think you're adding presumptions to what I'm saying that I don't intend. I mean selecting a single note is exactly that: selecting a single note. That is not a guess. That is literally what the user did, and the software should interpret that as their actual intention.
In reply to I think you're adding… by tromboneandrew
A note on single-note selections: I see the palette hairpin tool does, in fact, give the "correct" length if a chord is selected - for example, a half-note dyad followed by a half-rest in a 4/4 measure. But if that 4/4 measure has only a single note, not a chord, the behavior changes. Perhaps an issue here is that a dyad is selectable as a multi-select, which forms the blue box specifying a range, but a single note is not similarly selectable to create the blue box multi-select, if that is all that is in the measure: a half-note and a half-rest.
In reply to A note on single-note… by tromboneandrew
A single note can certainly be made a range selection. One easy way is to make sure nothing is selected (eg, press Esc), then Shift+click the note. Shift+click says "extend the current range selection from its current state, so that encompasses the selected single element as well as whatever it current encompasses". So, with nothing selection. Shift+click extends the current range selection from its current state (nothing) to a state where the current note is (also) encompassed.
Another way is to click, then Shift+right to extend the selection one note/rest to the right, then Shift+left to de-extend it that same amount. This is what I normally do in those fairly rare cases (eg, that 5% of hairpins I want to encompass a single note) where I need a range that encompasses a single note.
Really, though, most operations I might want to do on a single note do not require a range, because nothing about these operations inherently requires a start and end point the way adding lines - by their very definition -do. So, an ordinary click is sufficiently to select the note in order to delete it or change it duration, pitch, color, notehead, visibility, or to add an articulation or other symbol that has a single point of attachment rather than separate start and end points. Even copy/paste works after a fashion for individual notes, with special behavior to handle the case of notes being copied and paste between chords of different durations in a way that is designed to feel pretty natural.
In reply to I think you're adding… by tromboneandrew
In that case, we should simply refuse to add the hairpin, because a hairpin by definition does not attach to single notes, but to ranges: a clear start tick, and clear end tick.
Again, though, to me, usability is more important. If a simple algorithm exists that can correctly guess the end tick the user actually meant with a 38% likelihood of success, to me that's way better than doing nothing, or doing something else that we know for a fact won't be right but 5% of the time at most.
In reply to In that case, we should… by Marc Sabatella
Based on the justification you've presented so far, I would assume that it should make no difference to the length of a hairpin if a single note or a chord is selected. The image below is for reference: both done with the palette hairpin tool; the first with the single note selected (as that is the only option as far as I can tell) and the second with the dyad selected via the multi-select.
Also: I'm not sure where you're getting your notation examples. I have never seen this hairpin notation used, where a crescendo goes into the end of a note, but the crescendo is written into the rest after the note ends.
In reply to Based on the justification… by tromboneandrew
A single note is just the notehead. Only a chord has a duration (and in musescore charod has one! or more notes), a head has just a pitch and a shape,
So a single note is never a range, as it hasn't start nor end.
In reply to Based on the justification… by tromboneandrew
To expand on that.
...I just spent some time in the manual looking up definitions. It seems that what we have been referring to "select a note", is not what we think. Turns out that when we click on a note, all that is "selected" is a pitch. Duration, or any other attribute, is not selected. Click on a note and try and copy and paste it into another measure. Nothing happens. According to the manual, the proper way to "select a note" is to SHIFT+Click the note. Then when selecting a hairpin from the palette, it is applied to just that one note. So that make three ways to apply a hairpin to a single note.
So, of course, if we "select a note" and apply a hairpin, MuseScore has no idea what we want because we didn't select a note at all. The software defaults to a full measure. Whether or not this is the most common hairpin length doesn't matter in this case. Perhaps if anything needs to be changed, it's how "select a note".
In reply to So...I just spent some time… by bobjp
"the software defaults to a full measure" - only if you use the pallet. If you use the keyboard shortcut it defaults to the single "clicked-on" note. One of these must be wrong. Whichever is right should happen whichever application method you use.
In reply to "the software defaults to a… by SteveBlower
I'm not convinced either is wrong, as such. But I get what you are saying. I think it might depend on your workflow.
Scenario 1:
1. Click on note
2. Shift+>
3. Hairpin on single note
Scenario 2:
1. Shift+click note
2. Select hairpin from palette
3. Hairpin on single note
Scenario 3:
1. click on note
2. Select hairpin from palette
3. Hairpin from point of clicked note to end of measure.
As a trackball mouser, I'v never used the shortcut to add a hairpin. I have run into, and been irritated by, the scenario 3 problem. And yesterday I learned why it happens. And it makes sense. I can do scenario 2 without moving either hand from their main positions. I'd guess that 97.75% of the time I don't want a hairpin on a single note anyway.
It just depends on need and workflow.
The takeaway for me is that because scenario 3 is mostly undesirable, then I need to use the palette hairpin correctly. Rather than let MuseScore guess.
In reply to Based on the justification… by tromboneandrew
I'm not sure what you mean by "multiselect" here, but the second measure shown is exactly what I expect if you corrected selected a range - a blue rectangle around the half note chord - then clicked the palette button . It again has absolutely nothing to do with whether it's a single note or chord - it's all about what you've actually made a range selection or not. If so, you get exactly what you asked for. If you fail to make the range selection and instead only click a single note, you get what you see in the first measure. Once again, if you want to specify exactly a half note range - start point beginning of half note, end point just before the next note - you need to actually select that range. And that works the same for either the first or second measure.
I never said there were examples where the hairpin extended into a test. That's a red herring - replace the rest with a note and we'd be having the same conversation. If you have two half notes in a measure, and you fail to make a range selection but instead select only one of the two endpoints for the line, MuseScore has to either refuse to draw a line (because one point isn't enough to draw a line, not in music, not in geometry either), or guess a second endpoint. The number of cases where there were two notes in a measure like that, and a crescendo started on the first note and covered that note only - ending before the second note - was surprisingly tiny, around 1-2%. This was over hundreds of examples I tabulated.
In reply to I'm not sure what you mean… by Marc Sabatella
Obviously, we're not going to change each others' minds here. It could be just personal practice for me, or I'm just looking at different music - I do a lot of jazz orchestra, for example. In my own writing, I will often have hairpins notated covering only the exact rhythmic duration of played notes, and not extending into rests even if it "fills" a bar. This is to visually reinforce to the musicians how important it is to be crystal-clear on those releases. That convention you like where hairpins extend to cover rests ending the full measures is pretty much never something I will use, unless there is a phrasing reason where the phrase extends past the end of the measure. Different strokes and stuff, I suppose.
In reply to Obviously, we're not going… by tromboneandrew
To be clear: I never ever said hairpins should normally extend over rests. Not sure where you got that idea. Of course that's not what anyone would want. On the contrary, I specifically said rests were a red herring - meaning, they are irrelevant to the discussion. The exact same results would be obtained if those were notes instead of rests. In the cases where you ask MuseScore to guess where you want the hairpin to end (because you neglect to specify this by making a range selection), MuseScore makes the same guess regardless of whether the measure contains notes or tests. It happens to be correct very often as I said, because in many real world cases, there aren't rests - the measure is full of notes. That's what happened in literally over 30% of the hairpins in the published scores I consulted. In the cases where there are rests, indeed, it will always be the wrong guess. Sure we could improve the guessing algorithm to stop at the first rest, but the point is, it would still be just a guess, bec ause you failed to tell MuseScore the endpoint.
So again, the point is, don't ask MsueScore to guess. Simply make the correct range selection and you get the correct result, each and every time without fail. Want a hairpin extending over one note? Then range select that one note., Want one extending over a full measure? Then range select that full measure. Want one extending exact three notes within the measure? The range select those three notes. Want over extending over six measures? Then range select those six measures.
In reply to To be clear: I never ever… by Marc Sabatella
You referenced counting hairpins extending over rests in historical examples in a couple of posts here.
Because of this conversation, I noticed that in the "lines" palette selections, most of them do have this same behavior: with a single note selected and not a multi-select, the line object automatically extends to the end of the measure when selecting it from a palette: 8va, pedal markings, trills, but not slurs and ending markings which sensically have different behaviors. So, at least that's consistent. I don't think I'll ever understand why you insist on assuming a "guess" - my assumption is that if I select a single note, that means the rhythm of whatever "line" I want is defined by that note duration. Whatever, though. Obviously, many people here really like being able to auto-fill those lines to the end of a measure.
Heh, speaking of which: this could be an activate-able option somewhere: to autofill lines to end-of-measure or not. Then, the different behaviors between the palette selection and the keyboard shortcut selection could be made moot.
In reply to You referenced counting… by tromboneandrew
It's hard to be clear using words to describe music, so apologies if my meaning was not coming through correctly. I'll try again to clarify with some pictures this time.
But first, let me try to clarify something with words. A line by definition needs a start and endpoint., This is basic high geometry. If MuseScore is going to draw a line in your score, it can't do this given only a start point.
Everything about the definition of a line requires a start and endpoint. That is why I keep pointing out, selecting a start point only forces MuseScore to guess what endpoint you mean. You're not supposed to select a start point only, you're supposed to select a range. Then you have given MuseScore the start and points, and now it knows exactly how long to draw that line.
In cases where you select only a single note, MuseScore knows the start point for the note. Yu seem to be saying, since no end point was specified, we should calculate one for you based on the guess that you intend the end point to be, just before the start of the next note - in other words, if you made a range selection that was as long as the duration of the single note. And indeed, such a thing would be possible. But what I'm saying it, usability-wise it would be less desirable than instead making a guess that is actually statistically over 10 times more likely to be correct: extending to end of measre.
So for instance, consider the following measure:
If you want to add a hairpin in this measure, absolutely the crrect way to do it is to select the exact range of notes you want it to apply to - whether that range is one note, two, three or four. So if you want only a single, simply range-select that single note:
Then when you add the hairpin, it will only cover that single note:
But this is what I mean when I say, that case only accounts for well under 1% of the hairpins in published music. While any combination is possible - the hairpin might cover the first note only, or the first two, or of the first three, by far the most common case turns out to be, a hairpin extending to the end of the measure:
This is the type of case I meant when I referred to 38% of all hairpins in published music go to the end of the measure they started in. Obviously, if the measure contains rests as in your example, those particular measures would not be among these 38%. But as I said, the cases where the hairpin was a single note only was well under 1% of all hairpins. In other words, statistically speaking, this case here just doesn't com up nearly as often:
So while we would certainly prefer you don't make MuseScore guess - the correct way to add any line is to first make a range selection so no guesswork is required - we don't want to inconvenience the user by refusing to guess - by putting up a dialog saying "sorry, you only specified one of the two endpoints, so we aren't going to add the hairpin at all". Users don't like that. Instead, we make what is statistically proven to be the most likely guess: that the users wanted the hairpin to go to the end of the measure. It's still the wrong guess over 60% of the time (including each and every measure that ends in rests). So probably you'll need to adjust. Hopefully this will remind you to make the correct range selection first, next time. But still, having to adjust 62% of the time you forget to make the range selection beats needing to adjust 99% of the time you forget to make the range selection, which is what you'd need to do if we guessed you meant a range that only extended to the next note.
Aga8n, bottom line, please don't make MuseScore guess. Just select the actual range you want each and every time. But if you do forget and accidentally select only the start point and not the end, we do our best to do what it most likely to be the desired result most of the time. Given the original measure I am showing with the four quarter notes, by far the most common case for hairpins starting on the first note is, to end after the last. Nothing else even came close.
In reply to It's hard to be clear using… by Marc Sabatella
But still the puzzling question remains: why does MuseScore make a different guess depending on whether the keyboard shortcut or click on the pallet is used?
In reply to But still the puzzling… by SteveBlower
Steve, the most likely explanation I get from this conversation is that different people implemented each.
In reply to Most likely explanation I… by tromboneandrew
My next question then is why are there two implementations of one function?
In reply to Most likely explanation I… by tromboneandrew
I agree :-). And as I've said several times here, I agree this is worth fixing - but to me for all the reasons I've explained, the best fix to handle this specific case of user error is to go with the statistically most common likely scenario - extend to end of measure.
In reply to It's hard to be clear using… by Marc Sabatella
Marc: well, at least you seem to understand most of what I said, except you refuse to believe I mean what I say with your paraphrase "You seem to be saying, since no end point was specified, we should calculate one for you based on the guess that you intend the end point to be, just before the start of the next note"
In the case of the 4 quarter notes you included the image of here, my instinct would be to select the whole measure, not the first note. But now I understand you more: yes, I agree it is much more common in cases like that for hairpins to cover entire measures rather than just the first note or two.
Another thought: since we also agree it looks ridiculous to actually have hairpins extend through the entire measure when the measure ends in rests, an alternate ruleset for your "guessing" could be: fill the next consecutive set of notes until either the measure ends, or a rest is encountered, or another dynamic marking is encountered. For example, I was just looking through a Tchaikovsky score where the publisher very commonly ended diminuendos with piano and pianissimo markings at the end of diminuendo hairpins.
In reply to Well, at least you seem to… by tromboneandrew
Hairpins that extend over rests are quite sensible in piano scores when they may refer to a lower staff full of notes while the top staff has rests. Generally one would attach the hairpin to the upper staff to have it appear between the staves.
In reply to Hairpins that extend over… by SteveBlower
Steve: to clarify: I agree, but it is less common in many other kinds of scores.
Ideally, these rules would apply to all such "lines" (not just hairpins, if the Musescore team wants to stick with this concept): my last proposed rule "until another dynamic marking is encountered" should probably actually be "until another object of the same type is encountered" - hence, pedal markings, 8va markings, etc.
In reply to Well, at least you seem to… by tromboneandrew
It's not a question my refusing to believe anything. I get that for whatever reason, you wish it were the case that MuseScore guessed differently than it does in those cases where you neglect to select the full range. I absolutely believe you wish that. I am simply explaining why, statistically speaking, it would be sub-optimal for MuseScore to behave that way. I'm interested in saving the user work by doing the thing that will require the least amount of adjustment after the fact.
But, yes, I do agree a heuristic that looked at the content of the measure checking for rests could be better still. Although it gets hairy if multiple voices are involved, or multiple staves as mentioned. In any case, it's all mostly moot: the real way to guarantee the results you want is for you to make the selection correctly, regardless of the actual duration you want the hairpin to cover. Then it's not right 1% of the time, or 38% of the time - it's right 100% of the time, because no guesswork is involved.
So again, if you want a one-note hairpin, just make a one-note range selection.
In reply to It's not a question my… by Marc Sabatella
To answer Steve's question, there are different ways to do most things in MuseScore. Entering notes for one thing.
But more than that, don't forget that clicking on a note with the mouse only selects the pitch. Nothing else. As Marc says, you need to completely select the not to enter a hairpin, or other line, using the palette.
In reply to To answer Steve's question,… by bobjp
That doesn't answer my question. My question is: whatever the rights and wrongs of the outcomes why, if I do the allowed action of selecting a note or pitch as you refer to it and the do
a) press "<" on the keyboard to add a hairpin.
I get a different result if I do
b) click on a hairpin in the pallet. ?
It is the difference between the results for actions a) and b) that puzzles me.
In reply to That doesn't answer my… by SteveBlower
I understand. It is a puzzle.
But:
a) Select note (pitch)
b) Shift + >
c) Result - single note hair pin
a) Shift + select note
b) Shift + >
c) Result - single note hairpin
a) Shift + select note
b) Select hairpin from palette
c) Result - single note hairpin
Why does 1 work at all based on what the manual says about note selection?
In reply to I understand. It is a puzzle… by bobjp
Indeed, If The Way to do it is to select a range first, why should MuseScore do anything if you don't select a range. Why set up the user to experience puzzlement?
In reply to Indeed, If The Way to do it… by SteveBlower
I didn't even know that I could add a hairpin with ">" until a few days ago. But as I don't use shortcuts for the most part, it isn't a puzzle for me. It breaks my workflow. Besides, I I can't think of a time I put a hairpin on a quarter note. But that's just me, I know.
In reply to Indeed, If The Way to do it… by SteveBlower
It would be reasonable, but I think, not very efficient. I'd personally rather have it do something close to what I meant rather than nothing. I like efficiency, and I think most users do. If MuseScore can make reasonable guesses, I'm all for it.
In reply to That doesn't answer my… by SteveBlower
As I've already answered many times, I agree, this inconsistency is not good and should be fixed, by making both operations do the more common/useful thing.