Hairpin position changes on reload if horizontal offset applied on small staff

• Feb 2, 2019 - 12:47
Reported version
P1 - High
S3 - Major

I'm working on a project in Musescore, and it includes a system of two small staves (separated instruments) and two normal staves of a piano below them.
As I have to save space, I made the space between the staves of each instrument smaller. For most of the dynamic hairpins, I canceled the auto-placement so I could manually move the hairpins of the small staves closer to their parts (I also set the vertical offset of 1.5 as style).
But it seems like no matter how many times I reposition them, they change positions and even lengths every time I exit Musescore and start it again.
It's very frustrating, since not only they can't stay at the positions I need them at, but they also collide with other elements such as slurs, notes etc.


Status active needs info

When you turn off auto placement, items should stay where you put them rather than repositioning themselves. That is a bug. That you're turning off auto placement is an error on your part. Once an item's auto placement is turned off, it becomes the user's responsibility to keep it from colliding with other items, MuseScore's anti-collision algorithm doesn't take it into consideration.

If you want an item, like hairpins, to default to being closer to the staff, then edit the default offset in the Format->Style... menu so they CAN be closer to the staff. I think hairpins and dynamics default to being too far from the staves and make this adjustment in all of my scores by saving the style once then loading it in each score.

A sample demo score would be nice to demonstrate what you mean about hairpins moving on reload, so the actual bug can be fixed.

In reply to by mike320

thanks for the advice!
here is the score of the project that I was talking about, and also a screenshot of the first pages, that shows how I organize the hairpins every time I start Musescore, so you could see the differences.
Almost every hairpin is manually adjusted, because auto-placement wouldn't let me move the hairpins as close to the small staves as I needed it to be.
פרוייקט גמר.mscz screen.jpg

Attachment Size
פרוייקט גמר.mscz 61.13 KB
screen.jpg 700.71 KB

There are definitely problems. I changed the hairpin position below Y value to 0, pressed ctrl+a the ctrl+r to undo all manual adjustments you made. There are a few issues that should be addressed, like the hairpin in (reported) measure 2 shrinking back to the default every time you close and reopen the score. Another example is the hairpin avoiding the 5-let by too much in measure 12.

The need for turning off manual placement for the occasional item will never be eliminated, but should not be a standard practice. There are far too many tools available and the purpose of auto placement is to make your score entry easier.

Attachment Size
פרוייקט גמר_0.mscz 48.39 KB

In reply to by mike320

Workaround Yes No

It's geting extremely worse now, and the hairpins that are auto-positioned are changing as well.
פרוייקט גמר.mscz
See small staves measures 2,4,8,14,16,20,21,34 (only changed position until now. now gets shorter and shorter as well),35,37 (gets shorter),38 (gets longer), 39+40 (both get longer), 40,48,54-57 (all three),62,64 (last beat to first beat at the next. same as with measure 34), 79-80 (same as with measure 34),81,98,99.
I just started adding slurs and hairpins for the next measures, so I don't know the effect on them yet.
Is there anything I can do about it so I can save the positions as they are? Or even stop it from getting worse every time I open the software?

The offsets and length are very important, not only because it has to be possible for the performers to read, but also because it is a huge project that I make within the classical music major at my school, and it affects my final grade. I would transfer it back to Musescore 2, but it can't read any files from Musescore 3. So I'm really stuck with this bug.

Also, should I change the bug description to "active"?

" I would transfer it back to Musescore 2, but it can't read any files from Musescore 3."
Export in XML format, and version 2 will be able to open it.

In reply to by yuvalullman

Can you provide a single clear step by step procedure to reproduce the problem? Issues with hairpins sometimes moving after a save/reload or other operation are unfortunately not unique to 3.0, I've fixed quite a few in 2.x as well, and some indeed remain. I am not aware of any new such issues in 3.0 though.

In general, it shouldn't be necessary to disable autopalce just to move a hairpin - you should be able to move it until it comes within 0.5sp of something. But to get it closer, you do indeed need to disable autoplace.

In reply to by Marc Sabatella

Well, I guess if someone wants to recreate it, they should've simply uncheck the auto placement of a hairpin and move it closer to the staff than the minimum distance auto placement can handle. Next time they will enter the software, they will find that the hairpins are not at the same place and length as they were set in the last time.
I had to move the hairpins closer than the auto placement to the staves, to save space, so I unchecked the auto placement for the hairpins that I needed closer to the staves.
As I added more hairpins to the score, it just got carzier and crazier every time I entered Musescore. The last link to my score that I sent here shows exactly what I mean. And it wasn't only the manually placed hairpins that moved all over the place, but EVERY hairpin was.

I tried setting a smaller vertical offset as style. Nothing helped.

I just tried it and didn't have any problems, So it must be something pretty specific to the particular score and the particular hairpins or particular procedure or whatever. So again, a very clear series of steps to follow that is guaranteed to reproduce the problem would be helpful. As it is, I cannot reproduce,

Title crescendo and diminuendo hairpins change their shapes and places every time I start the software Hairpin position changes on reload if horizontal offset applied on small staff
Type Graphical (UI) Functional
Priority P0 - Critical

I did, but I only tried small horizontal sifts and larger vertical ones. I have now found I can reproduce the problem from scratch by using a larger horizontal shift. So here is a simple case:

1) load attached score
2) drag both hairpins left to appear to align with previous notes
3) save
4) reload

Result: the bottom one is fine, the top one on the small staff is misplaced

A possibly related aspect: the hairpin on the small staff should be small but isn't. Probably we are failing to scale things in a couple of different places.

Attachment Size
hairpin-shift.mscx 8.96 KB

In reply to by Marc Sabatella

Regression Yes No

I have found a similar issue, also related to hairpins and small staves. In the attached score, try to move all the hairpins for the small staff as close as possible to the staff (without switching off the auto-placement). Then, save the score, close and reload it. You will find at least one or more hairpins back to the "far" (default) position.
The problem is even worse after creating the part for the clarinet.

Attachment Size
Brahms - Nachtwandler.mscz 34.62 KB
Fix version