Able to directly input note/rest at every beat

• Nov 20, 2019 - 05:33

Hover above the 2nd/3rd/4th beat (a virtual image is shown on that beat) of a 4/4 empty measure in input mode, click and that note/rest is attached to that beat, instead of always starting from 1st beat. This can significantly save time. Sibelius has it and I really think it's very efficient.


In reply to by SteveBlower

I have no idea ow to do it with the keyboard. Using the mouse, you can place a note after an 8th rest. First you need to change Snap Positions to an 8th note in Preferences as I stated above. plus, after that change, you can place a note on any 8th in a blank measure.

In reply to by bobjp

So you have to change the snap positions rather than change the note length (that was what you actually said). And if it has to be changed in "preferences" that seems likely to involve rather a lot more clicking than just N, 3, 0, 4, G.

Still, one gets used to the way things work. I converted from another notation programme to MuseScore about 5 years ago and after a bit of fumbling around I have found MS note entry to be simple and fast to use from the pc keyboard (mousing is much slower). However, I am back to fumbling again after accidental entry has been changed from post-fix to pre-fix. But I will get used to it soon enough I hope.

I am all for other note entry methods, but wonder if what is proposed offers much gain. It seems just to have a set of different pros and cons.

In reply to by SteveBlower

I haven't been using Sibelius for quite a long time, or more exactly, I haven't used Sibelius for more than perhaps two hours, so I didn't notice the details of changing preferences in it. But anyway, this discussion topic isn't at all about preferences, I'm sure Sibelius can input a 8th rest and then 16th note just like MuseScore without any preference changes.

Let's just say only entering on an integer beat is allowed, still, the way I purposed is generally a lot easier, even if it isn't in some cases, it certainly cannot be harder (slower). If you want to input notes after 0.5 beat, there's no difference; but if you want to input notes after 2.5 beats, you only need to put a 8th rest on the third beat and input what you need, instead of inputing a minim AND a 8th beat before inputing notes, which is in fact very clumsy.

You may think it isn't a problem using keyboard to input, but for us who prefer mouse clicking (I just cannot find any musical sense of those relations between keyboard numbers and note durations, like 3 for 16th note, 4 for 8th note, 5 for ..., and it's another interruption to think about all these), this is what really matters when it comes to saving time and brain.

In reply to by Marc Sabatella

True, but imagine having a score with lots of 3/4 measures of a crotchet rest and minim notes. It isn't rare, I encounter measures like this in waltzs for orchestra, concertos, and others, and I'm really not a fan of entering all these rests first to allow entering subsequent notes. Still, what harm can it do to get note input much faster as explained? I'm not saying other issues are unimportant, but I think it isn't exaggerating to say that this can be the main concern of MuseScore 4 just like auto placement for MuseScore 3.

In reply to by Howard-C

I don't see this as important myself, it's rather easy to press the 0 or right click to enter a rest. If someone implements a method like you suggest then that's fine with me as long as it doesn't affect the way people have been entering scores til now.

In reply to by Howard-C

It just seems exceedingly strange to me to expect to not have to enter a rest on beat one when, if it had been a note on beat one instead of a rest, you'd never expect some sort of special mode to avoid having to enter t.

Anyhow, I'm not totally opposed to seeing this, but it seems like a lot of implementation work with considerable support implications (like, making sure the system doesn't break every time we change something else about even processing or layout) as well as potential inconvenience to people accustomed to entering notes and rests the normal way unless this behavior is made optional, in which case it's extra UI clutter as well. And I'm guessing that in practice it will only help in those really simple cases like your example above, and in most real music you'll constantly be fighting MuseScore for control over what beat to enter your note on. So it seems like an awful lot of downside for very little benefit to me.

In reply to by Marc Sabatella

It can be made optional.

It's just I have been wondering why do a lot of people around me choose Sibelius instead of MuseScore, and under comparison, these details really matter. Almost instantly when I gave Sibelius a try, I could understand why it is so convenient in many ways. MuseScore is more convenient in many other ways too, like the UI design, but why can't MuseScore keep the absolutely good things while making other things better? ;-)

In reply to by Marc Sabatella

I now have three in mind:

  1. This;

  2. Make auto-generated rests rhythmic, for example enter a crotchet in a 3/4 measure, following rests are two crotchet rests instead of one minim rest. More complicated case is, enter crotchet in 12/8 measure, following rests are 8th-4th-8th-4th-8th-4th-8th or 8th-4th-8th-2nd with dot or 8th-4th with dot-2nd with dot. There is toRhythmicDurationList() existing, but it doesn't always give desired results as discussed in another thread;

  3. Be able to re-layout measures by dragging barlines.

In reply to by Howard-C

It worked but I didn't like that there was no really direct relationship between measure stretch (which is what the adjustment was actually changing) and measure width (which is what you really want when doing this). That is, I might drag a barline, and think, oh good, now the measure is now the wdith I want, but as soon as something else changes in the layout, the width of your measure may change, just as is the case currently if you change stretch. What I think we really want is ability to set fixed measure widths. Which, frankly, would also not be that hard to do, we'd just need a setting for fixed width in measure properties and then use that instead of the calculated width in the system stretch calculations, but it has compatibility implications.

In reply to by Marc Sabatella

I think it's normal and acceptable to have measure widths still change when elements change after dragging the barlines, because since layout is supposed to be perfected after all inputs are done, if inputs are changed, the layout becomes obsolete and you need a new layout. If I'm going to drag the barlines I will wait until all elements are done input, and if I change something, I'm happy to see the layout getting along with it.

In reply to by Marc Sabatella

I wonder if the needs of an arranger might be different from those of a composer. For example: if you are arranging an orchestra piece from a 4 part hymn, you already have a blueprint to work from. As a composer, I move around a growing score. I absolutely often enter notes in various parts of a blank measure all the time. Sibelius is better by far for this kind of thing. I input notes with a mouse. If I want a quarter note on beat three of a blank measure, I move the cursor to beat three and click. By resetting the snap to reflect an 8th note, I can do the same at any of 8 positions in a 4/4 measure. Moving to MS from Sibelius, I sometimes forget the extra steps to do what can be done with one click in Sibelius.
Are these differences important? Maybe, maybe not. Although sometime they influence what I write.

In reply to by Howard-C

The issue isn't the autogenerating of rests, it the guesswork required. We can autofill rests upon entering a note if we know exactly what beat that note is being entered on and exactly how long that note is. And we are equally capable of doing that before or after the note. See, for instance, what happens if you have a single eighth note on beat three in voice 2 of an otherwise empty measure, then move it to voice 1 using the voice button. We correctly fill the rests both before and after it.

So, we already know how to autofill rests in either direction. What we don't know is how to guess which beat you want to enter a note on when you click in a measure. The current algorithm is completely predictable and fast - notes are entered left to right, or directly on top of existing notes or rests. At no point does MuseScore need to guess what beat you mean. It's that guesswork - and the complexity of developing a whole new subsystem that would no doubt have ramifications for the existing input processing system - that is the cause for concern. Not the generating of rests, which is trivial and already solved.

In reply to by bobjp

As a composer, it's difficult for me to imagine a scenario in which I somehow not know that a note on beat four needs to be preceded by three beats of something, or that my thought process would somehow change if those three beats are to be filled with sound or silence. Three beats is three beats, I enter those three beats left to right just as surely if they are notes or rests, just as surely if I am composing, arranging, or transcribing. I read music left to right, and I write it the same way.

In reply to by Marc Sabatella

"Not know that a note on beat four needs to be preceded by three beats" is not what I'm talking about. It's just on many occasions I feel like entering a note on, let's just say beat four, because the note on beat four is connected with following measures, it and following notes form a musical sentence, it doesn't actually have relationship with the rests before. If the note were in the first measure, that measure would only have one beat. I don't know how it's said in English, just "incomplete measure"? But I believe you get what I mean, and I feel it uncomfortable to have to firstly input a bunch of rests which don't belong to the musical segment I'm about to write.

In reply to by Howard-C

All I can say is what I said before, I know those beats are there, I can't force myself to not know, I read it that way, I can't imagine not wanting to write it that way - anything else feels completely unnatural. Maybe if I had never read or written music before but only dealt with MIDI or audio applications I might think differently.

In reply to by Marc Sabatella

So you are saying I have never read or written music. Nah, I'm kidding. I will take one step back too: it is not that I don't know "the beats are there", but one word: tedious. This is the way I feel for the current mechanism, especially when I'm dealing with this situation repeatedly like transcribing string parts of a concerto.

In reply to by Howard-C

No, I didn't say anything about you. I am simply talking about how I think, because someone speculated "I wonder if the needs of an arranger might be different from those of a composer". They were thus implying that maybe others who are composers would think similarly to how they do. So I explained how, as a composer, I am personally going to continue to think in terms of written notation when I write notation, because I can't very well not think that way.

As for tedium, to me tedious is having to use the mouse at all. It's hard for me to care too much about how to make mouse entry slightly less inefficient than keyboard entry. Whether it's five times more tediuous or only four is not an interesting question. But even if someone did force me into the tedium of using the mouse, it would be even more tedious to have to change my thinking from the normal left-to-right to something else just because there are rests at the beginning of a measure.

In reply to by Marc Sabatella

If you use the keyboard then you certainly don't care, but there're still mouse users. Personally I don't very much like the idea of firstly translate input info into keyboard language before inputing.

And my answer is: no matter what I'm doing, arranging, transcribing or composing, I always prefer to enter notes on whatever beat I like.

In reply to by Howard-C

Right, which is why I said from the beginning I'm not opposed to this, as long as it in absolutely no way negatively impacts anyone else, including others who use the mouse but prefer left to right input, and including the people who will need to continue to maintain any code added to support this.

In reply to by Howard-C

I think what Howard is interested in is the option to not have to enter rests first. It is an option in Sibelius. Marc, no one is asking you to change how you think or write. I use the mouse because that's how I learned, and I'm fine with it. To me, speed of input is not paramount. I know speed is important to others. As a composer, I am constantly stopping to ask myself if I really want this note in this instrument, octave and beat. As an arranger, those questions are not quite the same. Mouse input gives me time to think about the flow of the notes. What breaks that flow for me is having to stop to do the math to be able to put a quarter note on beat 7 of a 12/8 measure. It has nothing to do with reading left to right, and everything to do with how I learned to do the same thing in different software. In the other software do I miss sometimes? Sure, but it's how I learned. Just as you learned the methods you did. Some people have now problem learning to different methods. Great. For me, it's like playing guitar one minute, then picking up a banjo the next. I can't do it.
Same problem with drum set input.

In reply to by bobjp

I get that no one is asking me to change. I said from the beginning I'm not opposed to this if all my concerns can be addressed. Those concerns are valid and have nothing with whether my own workflow would be impacted. They have to do with how the change would affect other mouse users (not me), how it affects the complexity of the interface and the maintainability of the code, and whether the benefit (in terms of percentage of users who would actually welcome this) is worth the price.

I'm not saying the answer I definitely know the answers, but I am saying, if you want to convince anyone of anything on any of those points, you can't do it just by telling us how much you personally would like the feature. We already get that. You need to show there is a significant number of other users who would like the feature, and you need to show it could be implemented without affecting other mouse users adversely and without a significant impact on interface complexity or code maintainability. Howard is in a decent position to assess the code impact by actually doing a trial implementation, but ideally there would still be usability studies to assess how this would actually go over. Then we'd be better able to weigh the benefit against the cost.

Separately from that, I also addressed your speculation that maybe there is something inherent in the process of composition that makes one think differently about this. I pointed out that I am a composer as well and that I don't personally see the value in this for me. So it can't be that the workflow you describe is inherently something that composers will prefer.

For me, it really does have everything to do with reading left to right. I don't know that a note I am hearing in my head is on beat 7 versus beat 6 or 8 or whatever - I know what I would need to see written to get it to sound the way I want, That's because I don't understand rhythm by counting beats numerically, I do it by recognizing and reproducing patterns. I would have to do math to tell you that a note after two dotted quarter rests or a dotted half rest works out to be beat 7 in 12/8, but I could play it instantly by recognizing the pattern without knowing what beat it works out to.

So, obviously there is a difference between "people who think about rhythm like you do" and "people who think more like I do", but it isn't the same as the difference between composer versus arranger at all. Understanding what the difference actually is could be useful in deciding how to best design the interface around the feature, though.

In reply to by Marc Sabatella

Ah, now I see where the problem is. I'm not talking about left to right or rhythm (or thus a melody). i'm talking about a single note. The flute has an eight note pick up into bar 35 from a blank bar 34. What I'm used to is using the mouse to put that eight note on beat 4.5. Without having to do anything else in that measure.
It could also be a situation were the entire orchestra hits one note (say on beat three). You could just go down the score on beat three and have one action in each measure. I'm not saying all this is important, but the capability is out there.

In reply to by bobjp

Single notes have rests before them, so I read the rests left to right just as I would if there were multiple notes, that's how I understand what beat the single note falls on. If tat single notes is literally the last note thing in the measure, then sure, I can work backwards if need be, but I just don't read that way normally.

For places where the whole orchestra hits on the same beat, that's what copy&paste is for, or explode.

Again, though, it's not really that helpful for me to understand that you personally favor this approach. It's not going to make anyone else start favoring it, nor is it going to address any of the concerns I carefully laid out,

In reply to by Marc Sabatella

Marc, I'm sorry you misunderstand me. I should just let it drop. Howard asked a question, and I was trying to answer it. Of course notes at the end of a measure have rests before them. All I'm saying is that in MS I have to start a the beginning of a measure, add rests until I get to the place where I want to put a note. Perfectly rational. In Sibelius I go right to where I want the note and enter it. Rests are added automatically.
Sure copy and paste works for the hit except for clef and octave and different notes of a chord.
I don't know if anyone else works this way. But you might be surprised. I certainly understand concerns about cluttering up the code. I'll bow out as it's just not that important.

FWIW, if we're primarily concerned about adding a pickup, rather than having MuseScore try to guess which of twelve beats you want to place an eighth note on when clicking part way through an empty 12/8 measure, there could be a much simpler solution. What if there was just a command, "add note before cursor position", that worked exactly like the existing commands but it adds to the left rather than to the right? So, with eight note duration selected, doing this command (could be Alt+click, could be Ctrl+Alt+letter, whatever) while on the first note of a measure would simply add the note on the last eighth of the previous bar. No fancy guesswork or grids or anything else are required to figure out where you mean - it just subtracts the selected duration from the current time position and puts the note there.

In reply to by Howard-C

True enough. Allthough those aren't particularly good example either because they are so easily entered already. All the places where there are alternating eighth rests and eighths notes are just right click, left click, right click, left click, etc. And the passage where all the instruments have quarter rest, eighths note, eighth rest over and over is handled very quickly using copy&paste (and possibly repitch mode). Could also be done by entering the eighths as quarters initially (so it's back to right, left, right, left, etc) then shorten them all to eighths in one stroke. Either of these is going to be less error-prone than a system that relies on the precise position where you click to guess the desired metric position. Plus, getting used to taking advantage of copy/paste/repitch is going to pay off big time in cases where it doesn't just happen that the pattern being repeated involves rests. The copy/paste/repitch method works equally well whether there are rests involved or not,.

In reply to by Marc Sabatella

Easy? Yes, they were fairly easy before I came across Sibelius, but the moment I tried Sibelius, the convenience of note input immediately fascinated me and since then, I have continually been disappointed by the workflow in MuseScore. If we want to make MuseScore stay competitive, we should prevent users from having a feeling like that.

In reply to by Howard-C

Agreed as a general goal. But then we are back to the concerns I raised originally, which still haven't been addressed in any way. Namely, can this be done without impacting usability, complexity, or maintainabiltiy, and would this actually make a top 20 (say) list of pain points for a significant number of users? I want to make MuseScore easier to use just as much as you or anyone else does - but I would like to see the effort spent in areas where there is general agreement that it would provide significant benefit.

In reply to by Marc Sabatella

As long as it's made optional, users accustomed to the old way are unaffected, about complexity, I will have to dig into the code to tell (though I guess it isn't anything as hard as a GSoC project), and maintainability, I guess we can discuss this on Telegram or drag other developers' attention to this forum topic. I'm OK to maintain myself if necessary.

Well, do you already have a top 20 (or fewer) list in mind? If so, I'm happy to hear it, and help, if time and my mind allow me.

In reply to by Howard-C

As long as it's made optional...

If you're talking about check an option box in the properties (or somewhere similar) the design decision so far has been to not do this. If you mean a user doesn't have to do it if they don't want to, that's how everything works.

In reply to by Howard-C

Making it optional means adding an option, which adds interface complexity and potential for confusion (witness the number of questions here from people who have invoked insert mode and can't figure out why their measures have too many beats). Too many options take away from usability by making it harder to find the ones that actually matter and contribute to the very frustration you describe. One of the things people value about MuseScore over Sibelius is its relative simplicity. Bottom line, simply saying "make it optional" doesn't automatically solve the problem, it just creates another tradeoff to consider as part of the bigger picture.

Anyhow, I don't have a particular top 20 list in mind. Although it was interesting to see the comment earlier "I have been wondering why do a lot of people around me choose Sibelius instead of MuseScore". Without knowing the people around you but knowing other people who choose Sibelius, I can say they fall into one of three camps. 1) students whose teachers use Sibelius and tell them they must use it too, or 2) users who need some fairly esoteric layout feature we don't provide, or 3) users who want more control over playback than we provide.

For 1), there is little we can do but educate the teachers, or rely on the eventual groundsweel of opposition from students (both of which do make a difference, and quite a few schools are recognizing the value of MsueScore now). For 2), there is a pretty long list of layout improvements to be found by searching the issue tracker but probably the biggest are in the areas of general page layout - finer controls over staff size and spacing, instrument bracketing and group naming, etc. For 3), it's probably mostly about VSTi support.

In reply to by Howard-C

Yes, we all want MS to be easy to use. But what does that mean? The three people on this thread all have different approaches to using MS. What is easy to use for one isn't for others. I think it might be very hard to gauge what will will benefit the majority of users. There are people who come on the forums almost daily who have no idea how to write notation and how to use MS. Bless their hearts. Of course MS needs to be easy to use. There are plenty of things about MS that I will never use. They don't benefit me at all. That doesn't mean they shouldn't be there. This whole idea might not be possible. So be it. It might seem like an insignificant detail. But it might be the attention to detail that make MS attractive. Sure MS is free. People tent to not expect much from free software. The reason I'm using MS more is because playback is so much better than before. That's pretty much it. There have been many other advancements. They don't mean as much to me as decent playback. I know that, in the end, playback is down the list for many people. What's important is a way to transcribe something, or write something, in order to get music printout for people to play.

In reply to by bobjp

It is indeed hard to determine what will be the most useful, but that's why there are usability experts and processes for creating formal usability studies to help figure these things out. It's all too easy and common to assume that "whatever I find useful will be useful to others" and end up designing things that end up being useful to no one but the designer. It's also all too common for six different developers to each have six different ideas for "simple" changes that end up being kind of mutually incompatible because there was no one thinking about the big picture.

So again, to me it is completely pointless to argue about what any of the tiny percentage of MuseScore users on this thread happen to think would or wouldn't be useful. If one wants to make a significant difference in usability, you need to have better information than any of us can provide on our own.

In reply to by Marc Sabatella

I'm not asking for this ability. I'm just defending it's usefulness.
1. I understand it is only useful for mouse users.
2. There probably aren't many requests for it because almost no one know it is possible.
I care less about this ability than I do about other aspects of MS. What are the processes for determining usability studies?

In reply to by bobjp

But it's already possible to input on any beat if every beat already has note. You can choose to add to whatever note you want to form a chord, beat 1, 2, 3 or 4. And if every beat has a quarter rest, you can do that too. People only don't know it is possible to add note on every beat if there's only a whole rest, but if you (I) think about it, it's completely normal to enable that.

In reply to by Howard-C

I have been trying to work out how this entry method would work in practice. Based on my understanding of Howard-C's and bobjp's description I have mocked up the following screen shots, that show what I think would happen with a simple example.

Desired result:
01 Desired result.png

Score at start:
02 Start.png

Set snap granularity to 16ths and hover over 1st measure:
03 Set snap granularity to 16ths.png

Enter first note:
04 Enter first note.png

Enter second and third notes:
05 Enter second and third notes.png

Enter half rest:
06 Enter half rest.png

Enter quarter rest:
07 Enter quarter rest.png

Enter eighth rest
08 Enter eighth rest.png

Hover over second measure:
09 Hover over second measure.png

Enter quarter note:
10 Enter quarter note.png

Enter half rest:
11 Enter half rest.png

Enter quarter rest:
12 Enter quarter rest.png

If I have understood this correctly, one potential issue seems to be the way the space taken up by the measure being worked on has to change to accommodate the "ghost" rests that indicate the available positions for notes.

Another difficulty is working out which of those 16 positions you actually want to drop your note on. It is fairly straight forward in my example of a three note pick up but how about this case:

Desired result:
13 Example 2 desired result.png

Where do I drop the note???
14 Example 2 which position.png

I can see that in some limited cases it may be useful to those who prefer mousing to using the keyboard, but it also seems to make note entry with the mouse more difficult in other cases. And, while I don't have experience of MuseScore coding, my gut feeling is that quite significant effort would be needed to implement this, and in particular in dealing with the layout changes when "ghost" rests have to be displayed. I wonder whether it aids note entry significantly to make that effort worthwhile. Certainly if implementing this hinders or complicates the existing note entry methods, I hope that it would not go ahead.

In reply to by SteveBlower

It isn't the mechanism's fault if you set the accuracy too high, like 16th note. You're just messing with your own eyes.

And no, there shouldn't be sixteen short rests when you hover the cursor above, the rest will still be whole rest but the ghost note appears on a certain beat.

In reply to by Howard-C

Seems I didn't understand then. I am genuinely interested how this would work. Please can you help with a bit more explanation.

I don't see why setting an accuracy of 16th note would be too high if it is a 16th note that I want to enter, or rather if I want to enter a note on a particular 1/4 of a beat. But then perhaps I am misunderstanding.

What would you actually see when you hover that helps to place the note on a particular beat or part of a beat?

In reply to by SteveBlower

This is probably how it works under the hood. But Sibelius automatically converts all the rests to their lowest denominator. The effect is that the measure has a whole measure rest. With note input active, the mouse drags the snap-to ghost note to the proper position. Click the note in. Done. Nothing to clean up.

In reply to by bobjp

I don't know about keyboard note input, but the ghost note is always there with mouse input.

The pointer doesn't show in this example. Start with a blank measure and drag the note to the proper place. ghost.jpg

And click...

  1. select the note input value.
  2. drag to position
    3 . click.

In reply to by bobjp

But how do you know the note is in the proper place? In your first picture I see nothing that helps me see the note will be entered where it is in your second picture rather than a 16th earlier or later. Based on the first picture I might guess it would be entered somewhere during the third beat but exactly where within that beat is not clear. And if were dealing with a 32nd, the scope for error seems much bigger. Again I must be missing something.

In reply to by SteveBlower

It's hard to describe. My "snap-to" setting is eights, but it could be 16ths or 32nds. but that would be crazy.

What you can't see is that as I drag the ghost note around it jumps or snaps to the next position. Even if I miss the target beat, I can undo and try again. All of which seems faster to me than filling a measure with rests then entering a note and then fixing the resulting rests.
MuseScore does not remotely work this way. I'm not saying it should. I understand how someone proficient at keyboard entry would perhaps have no need for this kind of thing.
As a mouse user, this what I think I need to do in MS.
1. After I remember that I have to start at the beginning of the bar, I select a double dotted half note.
2. Select "rest"
3. Click the new rest into the beginning of the bar.
4. Enter my 8th note at the end of the bar.
Certainly doable.
In Sibelius, I click the note where I want in the measure. A 32nd note could be fraught with danger, to be sure.

In reply to by bobjp

Thanks. I think I understand better now.

I thought about how I might use mouse only to enter a three 1/16 note pick up a the end of an empty 4/4 bar. I would probably do something like this:
1. Select 1/2 note duration
2. Right click in measure - result two 1/2 note rests
3. Select 1/4 note duration
4. Right click on 2nd 1/2 note rest - result 1/2 note rest + two 1/4 note rests
5. Select 1/16 note duration
6. Right click on 2nd 1/4 note - result 1/2 note rest + 1/4 note rest + two 1/16 note rests + 1/8 note rest
7. Left click to enter three 1/16 notes at the pitches I want.

There is no need to fix rests as they are correct already.

What struck me about this was that most of the clunkiness came from the need to move between the measure and the note duration tool bar and that bit of clunkiness is there even if the placement of the entry point was controlled as you have been describing and I have been failing to understand (my fault!). One could use the keyboard shortcuts, but that defeats the objective of mouse-only entry.

I then wondered if the mouse wheel could be used to scroll through note durations. Unfortunately, the mouse wheel is currently used to control zoom and vertical and horizontal panning. However, it seems that both shift and alt are used with the wheel to achieve horizontal panning. I wonder if one of those could be used when in note input mode to activate scrolling through note durations. And of course after searching the Issue Tracker, I found that someone had already had a very similar idea #284833: Change note duration with mouse wheel . This seems to me to offer more scope for speeding up mouse-only entry and is probably much simpler to implement.

In reply to by SteveBlower

Or start a measure with a dotted half rest.
Enter 4 16th notes on beat four, the last three of which are the ones you want.
Delete the first 16th.

Even that is 5 steps.

One measure in one part might not be a big deal to save 2 or 3 steps. But it adds up in a big score.

In reply to by bobjp

There are many possibilities for choosing a set of rests to get you to the desired insertion point. The point I was making is that it is the actions needed to select the durations that make this process slow for me. And the reason for that is because of the frequent moves from the currently selected position in the score to the note duration tool bar and back again. My suggestion is that offering a method of selecting note duration that does not require the user to mouse away from the current entry location offers more scope for improving the mouse-only notation experience than the suggested method of putting the entry point at a variable place within a measure. Of course, both could be implemented but the duration selection issue seems likely to require less development and to be less likely to break something.

I came here looking for this exact feature request and I agree that this is something that feels really unhelpful when switching from Sibelius to Musescore.

In Musescore If you want to add a quaver at the end of an empty bar with the mouse, it's a real faff. The most "natural" way seems to be to add a dud note on beat 1, then another on beat 3, then another on beat 4, then the actual note you want, and then range-select all three dud notes and delete them.

Sibelius lets you input straight into beat 4 immediately, so you can just place two quavers and then delete the first one. What makes it just that little bit more annoying in Musescore is when the bar is wide and you are mousing around in the right hand side of it and Musescore is very obviously ignoring your intention and displaying a ghost note in beat 1.

In reply to by jeetee

Firstly, this doesn't actually work in the latest nightly build of Musescore 4. Right clicking gives you notes, not rests. I assume that's an oversight.

Secondly, unless i'm mistaken, this doesn't actually save you any steps. You just end up with a mess of quaver-length rests instead of quaver dud notes, which still needs cleaning up, for which you would do the exact same range-select and delete. So its the same number of inputs as if you were entering dud notes. No?

Or, you can switch duration as you go so that you are always entering the correct rest lengths, so there is no cleanup, but now the steps are: (1) select minim (2) right click beat 1 (3) select crotchet (4) right click beat 3 (5) select quaver (6) right click beat 4 (7) click beat 4.5 to enter the note. Hardly any better and I am still in the learning phase of mixing up the keypad keys for durations, so this is definitely slow going for me.

I can't see any way of doing it that is anywhere near as quick as what Sibelius does.

In reply to by brokenbeta

If I were to use the mouse (which I don't, I use the keyboard) I would do it as follows:

  1. Select duration half
  2. Select duration dot
  3. Right click rest
  4. Select duration quarter
  5. Enter note

I get that working with a shadow grid might save some time in some specific scenarios (wide enough measures and a large enough duration subdivision). Which means that you've at least taken some prior action to create wider than default measures. I also think it'll be a lot harder to click in a 1/16th on the "correct" position that way, than it'll be to enter a quarter/half note.
But all that aside I know there's been at least some thinking around how this may be perhaps implemented in the future.

In reply to by jeetee

(i actually meant a 1/8th note as the example, not 1/4)

I think even if you have assistance entering notes on crotchets/quarter notes it helps a ton. As you say, finer granularity just means more risk mis-placing notes. The big deal is just getting to roughly the right spot quickly. And yeah, the bars might have to be wider by default. But that would honestly look better anyway IMO. I just wanna post some gifs to emphasise, the difference in usability between Musescore and Sibelius feels pretty big to me.

The least bad way you can do this in Musescore that i can see:


The IMO much better way you can do it in Sibelius:


In reply to by brokenbeta

So in Sibelius you have to enter a note you don't want in order to be able to enter a note you do want. That seems clunky to me; whereas in MuseScore you enter rests you want followed by a note you want which seems more sensible.

MuseScore works very well from the PC keyboard which doesn't rely so much on fine motor skills to place notes in their correct time slot and pitch. So, from the Keyboard, N (to get into note entry mode) 6 0 5 0 4 0 G. With a bit of practice it is quite possible to enter notes and rests very rapidly by touch typing using the number pad for durations and the letter keys for pitches. But then, I am old enough to remember when PCs didn't have mice. In fact I am old enough to remember when there were no PCs and everything was typed onto punched cards and results were printed 132 characters per line on fanfold paper.

In reply to by SteveBlower

No, you definitely do not HAVE to place dud notes in Sibelius just to get a rest. You can just type out rests as you do in Musescore. But in this case, it is definitely quicker to use the dud note, as it places several graduations at the exact spots that you likely want. So why not take the shortcut?

It is even better if the note wants to be directly on beat 4. Then we are talking about click done. Compared to that, faffing around with rests feels very clunky.


Don't get me wrong, I do believe in the speed of keyboard entry. However, I use a graphics tablet as a mouse, and I can attest that you can be very fast at reaching a particular spot on the screen with high accuracy. So for me the fastest input will always be some combination of keyboard + mouse, which tends to weigh more heavily towards keyboard the more of a power-user you become.

I can't speak to Musescore's keyboard efficiency and I'm not really musically proficient enough to judge anyway. What I can say is compared to Sibelius, its mouse input feels slow and awkward to me, and this issue is part of the reason why.

In reply to by brokenbeta

I'm not sure what the confusion is here. In Sibelius, I can put a note anywhere with one click. After selection the note value first. But there is a good chance it is already selected. I don't want to have to go back to the beginning of the measure just to create rests. I just want to add the pickup note. That's it. Left to right is in mo danger here. It's not even a question of what is faster. I just want to put a note where I want.

In reply to by bobjp

But that's part of what I'm trying to understand; In the example above Sibelius also doesn't let you put the note where you want. You can't add the 1/8th at the end of the 4/4 measure directly either, you need to first enter either a rest or a dud note on the 4th beat itself.
It gets you closer to where you want, and I do believe that that is a Good Thing(tm).

What I'm trying to understand is that when such a thing is designed for MuseScore it'd make sense to use the selected duration as a subdivision (which in this case would let you place the note directly where you want it) or whether that would become to imprecise.
It'll still not cover all scenarios as well: in case you need to put a note on a subdivision of the selected duration. If you want to put a 1/2 note on the 2nd beat of a 4/4 measure, then using the selected duration as a subdivision is worse than using beat subdivision.

I also think that if you let the user select/change the subdivision you're likely no longer getting the speed gains compared to entering rests. (Select note duration, select grid subdivision, aim, place note).

So currently my view is that when (not if, but when ;-) ) MuseScore adds this, it should likely follow the effective time signature as a snap to grid subdivision. A 1/4th in 4/4 and 3/4, a 1/8th in 6/8 and 7/8, a 1/2 in 2/2, a 1/16th in 19/16 etc..

In reply to by jeetee

Yes, so another way of looking at this is, Musescore will only let you start note input where there is a musical symbol. If you have an empty bar, this means you can only start on beat 1, which might be quite a long way from where you actually want to be.

Sibelius seems faster to me letting you start on 1/4 crotchet intervals since you are usually pretty close to where you want to be, but that doesn't mean I am married to starting at 1/4 intervals. I just want it to be faster to get to the right spot one way or another.

Meanwhile, here is Dorico's approach -- -- which uses a subdivision selection in the bottom left corner to decide how granular the steps are for entering notes and moving the caret. It doesn't use the time signature, it has a separate setting for it. I haven't actually tried Dorico yet, but this looks like a great solution to me, too.

In reply to by jeetee

There is a setting for mouse input called Snap Position. The default is quarter note. This why the above video uses a dud note to get a quarter note on the and of four. Mine is set to eighth note. As I move the curser in a measure, the ghost note snaps to eighth note positions until I'm at the end of the measure. Or whatever position I want. The Snap Position can also be set to several other values. It doesn't make any difference what the time signature is. Select duration, aim, place note. For mousers, this is faster.

But is it faster than keyboard input? Of course not. We mousers are always told that if we would just learn to use the keyboard then everything would go faster. It's all about faster. I get it. If I were doing things like transcription, faster would surly seem more enticing. I use a trackball mouse which is easier, and faster, than a traditional mouse. It's not faster than the keyboard, but then I don't have to remember anywhere as much. My brain doesn't work that way. Mouse input is easier for me. In the end, I just want to put notes on a staff. All this is just one of the ways MuseScore seems a bit less mouse-friendly. That is not a complaint. Just an observation. Of course different programs will do things differently.

In reply to by bobjp

+1 on note entry Snap Position for those entering notes by via mouse clicks.

And I agree, there's no need to opine with respect to the speed of mouse entry vs. keyboard entry. People will prefer different approaches, and sometimes choice their input method depending solely on the context or task.

Nevertheless some sort of "grid" input would greatly increase input speed for mousers. And the overall result would feel more like pencil and paper, which basically everyone understands.

In reply to by SteveBlower

As an aside, there's also the possibility of making use of the Piano Roll Editor functionality since it is by nature grid-based.

If the PRE in MS4 would automatically update depending upon user-selection across the score (including all staves), all the while being docked so that it doesn't lose focus, and then allowed for placing notes on that grid with a user-defined duration not contingent upon already existing score elements as it seems to currently be (Actually, you can do this already, just make sure to use "Insert Note" and not the default "Append to chord"... it already places rests automatically for you!), this would probably work pretty slick for the previously mentioned purposes. Hopefully MS4 can get that updated...

Also, the grid resolution should in my opinion be automatically updated when the user selects a particular duration within the PRE, or at least have a choice to have that be happening or not.

In reply to by brokenbeta

To be clear: right-click enters a rest. That's how it saves time - you don't need to first enter a note and then change your mind and say you really wanted a rest - just enter the rest directly.

In general, MuseScore expects you to enter notes and rests left to right. If you had a string of notes on the first few beats you'd expect to have to enter them manually. And if you were you writing with pencil and paper, and wanted a note on the & of 4, you'd need to write the rests, manually, Rests are as important to music notation as notes, so I would just get in the habit of not thinking of them as optional things you shouldn't have to enter in the cases where they happen to show up at the beginning of a measure. just work left to right and you didn't have to think about which rests you will need to enter yourself and which you won.t

That said, indeed, a feature where in certain cases MsueScore could fill in rests for you automatically could sometimes save some additional time, and I wouldn't be surprised to see that someday. Still, though, thinking and working left to right really does have its advantages when it comes to music notation.

In reply to by Marc Sabatella

It is true that on paper I would have to manually write the rests. I think paper is a good model for understanding what the score should look like at the end of the edit. If we are using paper as a model for a good style of input, well, I would say you have successfully jumped a very low bar!

There are already plenty of places in Musescore where rests are automatically filled in for you. Here is the most obvious one:


Where did all of those rests come from?! Can you imagine if Musescore didn't do that? It would be a nightmare! I don't even know how it would work. To help you out, Musescore presumes that any space at the end of your note input needs to be padded with rests up to the barline, so we do not cheat the time signature. We are all very happy that it does that!

Imagine if I argued to you that note input should NOT do this, but it's good because you'll get a better awareness of the remaining beats in the bar if you are forced to enter the rests yourself, as you would if you were writing on paper. I don't think anyone would buy that argument and I think you guys would all be looking at each other in confusion and wondering, "why does he want to make it worse?!"

In reply to by brokenbeta

To be clear, no one has ever said this shouldn't be considered as a way of making things even more efficient in the future. What we are saying is that, if you shift your thinking or approach to a more left-to-right model, you can leanr to work extremely efficiently eithin the current system.

And yes, MuseScore does fill rests now to the right of a note, because that's necessary in order for the left to right approach to work at all, for the reaosns you allude to. Most of the time you don't even intend to keep those rests - they are about the be overwritten by notes. S it's not really intended as a time-saving measure, even though occasionally - in the cases where your measure does indeed end with rests - it can have that effect.

Anyhow, indeed, there is no technical reason rests can be generated autoamtically where needed. Implementing a grid like this will no dooubt hapen someday. Meanwhile, though, it really is quite possible to work efficiently left to right.

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