Corrupted file because of weird tuplet
I'm transcribing a piece into MuseScore 3 and when I try to load it I get a message that the file is corrupted. The details given are:
Measure 178, staff 1 incomplete. Expected: 6/8; Found: 54/64
What I have in that measure is unusual but shouldn't cause corruption. I have a measure of 6/8. and starting on beat 4 I have a 16th note 'quadruplet.' The next measure goes back to 4/4 so this quadruplet is essentially "preparing" the feel back to 4/4 by acting like 1 beat (dotted quarter beat) = 1 beat (quarter note in 4/4), with 4 16th notes.
Any ideas how i can "fix" it so it's not corrupted? Attached is just that measure and the next one for reference. I also made everything visible because originally I had some stuff hidden.
| Attachment | Size |
|---|---|
| quadruplet.mscz | 9.4 KB |
Comments
Try the attachment (a rioty rest?)
Seee https://musescore.org/en/node/54721
I'm working on this in version 3.1-beta so I can see what might have been fixed or broken.
The first thing I notice is that the tuplet is all wrong. The measure is not missing a 16th note, it has an extra one. It looks like you used a custom tuplet to make 4 16th notes, and it should work fine.
When I recreate the measure the way that makes sense using a custom tuplet, it does not get corrupted in version 3.1-beta with close and reopen. It seems this problem was fixed by changing the ticks to fractions. I haven't seen any corruptions in the 3.1-beta. :)
In reply to I'm working on this in… by mike320
I noticed that the 16th rest really shouldn't be there because of the dotted eighth, so was that it? Also, are you saying this is an issue in 3 but won't be in 3.1?
Thanks!
In reply to I noticed that the 16th rest… by Bryan Saylor
No, 3.1 shows it the same as 3.0.5
In reply to I noticed that the 16th rest… by Bryan Saylor
The extra 16th rest is the corruption.
Opening the score in 3.1 does show the corruption. Making the same measure in 3.1, saving it and reopening it does not cause a corruption. I would prefer 3.1 to fix the corruption from 3.0.5, but I'm not sure that's a realistic possibility. Someone more familiar with the file and code would be more qualified to answer that.
In reply to I'm working on this in… by mike320
Huh? I see that tuplet missing an 8th
Hmm, it is not, it is a 4:6 tuplet, not a 'regular' 4:3 one, so indeed that 16th rest is too much
In reply to Huh? I see that tuplet… by Jojo-Schmitz
It should be 4 16th notes in the place of 6 16th notes, but with two voices: voice 1 should just be a 16th note followed by dotted 8th (as a quadruplet), voice 2 is 4 16th notes (as a quadruplet). I recreated it the same way a few times, making a custom tuplet as 4/6 ratio, but it adds an extra 16th rest at the end of the 1st voice.
Unless I'm creating it wrong I'm not sure what is incorrect about it. Below is how is "should" look (i've hidden some things):
In reply to It should be 4 16th notes in… by Bryan Saylor
And it does look that way when I enter it (except for the tuplet bracket)
In reply to And it does look that way… by Jojo-Schmitz
...and you used 3.1 like I did.
In reply to ...and you used 3.1 like I… by mike320
Looks and works the same in 3.0.5
Edit: no, darn, it does not!
So a bug fixed by the fraction changes in 3.1, good ;-)
In reply to ...and you used 3.1 like I… by mike320
Thanks Mike,
Just saw your previous comment and this one - so it is an actual issue, but will be addressed in 3.1? If that's the case then I can deal with it.
In reply to Thanks Mike, Just saw your… by Bryan Saylor
Your score would still be corrupt there, but entering it your way again won't cause a corruption anymore, by the looks of it