MuseScore 3 keeps crashing

• Aug 18, 2020 - 17:37

So I open MuseScore 3 as normal, after the recent update. Everything goes normal, until it loads up. Once it does, it crashes right after. No scores appear, even if I press yes on restore session.

Comments

• Jan 15, 2021 - 02:59

By new update, do you mean 3.6? Are you starting from the program icon or by double-clicking an existing score? What version were you updating from?

If it crashes starting from the program icon, try uninstalling, rebooting,. then reinstalling - it could be some file failed to update properly because it was in use by another process.

If it crashes starting by double-clicking a particular score, try starting from the program icon. If that works, but it then crashes opening that same score from the File menu or Start Center, please attach the score so we can inverstigate.

Reply

• Dec 11, 2021 - 04:45

Hi. I have the similar problem. MuseScore 3.6.2. (the latest) for MacOS. It has been working fine. It crashed a few time while I was editing a score. Then, it started crashing every time I try to "rewind" or "play" the score. I deleted, reinstalled, or factory-reset the MuseScore, also tried both from the score icon or App icon. None of that fixed the problem. Further, it crashes only on this particular score, not the other scores of mine I tried. I have attached the score (still unfinished checking its fidelity to the original sheet), as you suggested.

Attachment Size
Skye Boat Song.mscz 37.57 KB

Reply

• Dec 11, 2021 - 08:59

Confirmed, crashes on Windows 10 too, it is a different issie than what's been here reported earlier though

(BTW there's an "h" too much in the copyright test, "Publich Domain")

Stack trace:

1 Ms::RepeatList::updateTempo repeatlist.cpp 162 0xa62ba7
2 Ms::RepeatList::unwind repeatlist.cpp 991 0xa660d2
3 Ms::RepeatList::update repeatlist.cpp 143 0xa62aea
4 Ms::MasterScore::repeatList score.cpp 1830 0xa73a28
5 Ms::Seq::getPlayStartUtick seq.cpp 1210 0x4767d0
6 Ms::Seq::canStart seq.cpp 332 0x472638
7 Ms::ScoreView::::operator()(Ms::ScoreView *, const QByteArray &) const scoreview.cpp 2177 0x64288f
8 std::_Function_handler>::_M_invoke(const std::_Any_data &, Ms::ScoreView *&&, const QByteArray &) std_function.h 297 0x65de80
9 std::function::operator()(Ms::ScoreView *, QByteArray const&) const std_function.h 687 0x10ca20d
10 Ms::ScoreView::cmd scoreview.cpp 2890 0x64acd1
11 Ms::ScoreView::cmd scoreview.cpp 2089 0x6423b7
12 Ms::MuseScore::cmd musescore.cpp 6617 0x433e30
13 Ms::MuseScore::cmd musescore.cpp 6044 0x43077a
14 Ms::MuseScore::qt_static_metacall moc_musescore.cpp 610 0x5eedfd
15 void doActivate(QObject *, int, void * *) 0x68b978bb
16 Ms::ScoreTab::actionTriggered moc_scoretab.cpp 226 0x60e268
17 Ms::ScoreTab::qt_static_metacall moc_scoretab.cpp 111 0x60dcd7
18 void doActivate(QObject *, int, void * *) 0x68b978bb
19 QActionGroupPrivate::_q_actionTriggered() 0x32704cc2
20 QActionGroup::qt_static_metacall(QObject *, QMetaObject::Call, int, void * *) 0x32705299
...

Reply

• Dec 11, 2021 - 11:27

Hi. Thanks for that. I much appreciate it. As you said, the problem was 1.2.3.4. I have changed that to 1,2,3,4. That stopped the crash. I also changed "D.C. al Fine" to "D.S. al Fine". Now the score plays correctly. The original printed score did not use 1,2,3,4 nor 5. but I thought they make it easier to understand. Also, the last bar must have repeat count of 5. I am somewhat puzzled by this. Because, it seems more intuitive to let the last bar line (which is the repeating bar) to own instruction for repeat count. Is there some reason for this? Maybe it is to make it consistent with the use of other repeating notations?

Attachment Size
Skye Boat Song.mscz 37.01 KB

Reply

• Dec 11, 2021 - 11:32

Every time you want a repeat played mode than twice, you need to set the play count of the measure containing the end repeat barline accordingly, that's just the way MuseScore works.

That these voltas do work (if setup properly) is an added benefit, as normally voltas are alternative endings, not meant for the middle of a repeted section.

I'd have some code ready that prevents that particular crash, but I'd rather pass that on to @jeetee, our repeats wizzard ;-)

Reply

• Dec 11, 2021 - 16:25

@yoichi123...

You wrote:
the problem was 1.2.3.4. I have changed that to 1,2,3,4.

To be clear: In the Inspector, only the Volta 'Repeat list' must be comma delineated. The 'Begin text' did not have to be changed from periods to commas. Periods (or any other text) wil work.

You wrote:
I also changed "D.C. al Fine" to "D.S. al Fine".

I guess you did not want the piano intro. played each time, so did not not want Da Capo (to the head) but rather D.S.(to the Segno).
So... after the piano introduction...
Since your score has 4 verses with a chorus played once at the start, then once after each verse, then once at the very end, a simpler way of notating this would be:
Skye Boat Song_2.mscz

(Some piano accompaniment was removed for clarity in viewing the score)

Reply

• Dec 11, 2021 - 21:43

Thanks for that. I didn't use Volta list at this time in the inspector and did not realise that until now that it was the real cause of the problem. Despite that that was the clear advice I got from Jojo-Schmitz. Though, looking back, I did change Volta list in the past. As to your revised score, it applies Fine only at the 5th repeat. I guess that is your point. To me, however, that is rather strange. That is not intuitive. It means, D.S. al Fine applies only at the 4ht repeat and Fine applies only at the 5th repeat. The repeat structure of the score can be written in different ways, but the way I did was to copy the original score and keep the structure as it wanted. The point of the original score is that the final note "cry." of the Chorus has an additional length to keep when the song ends. The way how it is written makes it clear that the extra final measure (marked with Fine) is only played at the 5th repeat and that is where the song ends.

Reply

• Dec 11, 2021 - 18:14

@yoichi123...

You wrote:
Also, the last bar must have repeat count of 5. I am somewhat puzzled by this.

Regarding the "counting" of the number of play backs (play count) when using 'begin repeat' and 'end repeat' barlines...
A simple end repeat barline has its 'play count' (in the bar properties) automatically set to 2 (for a one time repeat), therefore any requests to return to the start repeat barline again (to play more repeats) must increment this initial value of 2.
Consider it this way:
When playback first encounters the start repeat, play count equals 1. When it gets to the end repeat, it returns to the start repeat (now play count 2). So, to further add one additional repeat, the play count must be incremented to 3: the playback engine initially plays the section (play count 1), then repeats (play count 2), then repeats again (play count 3).
For voltas...
If a volta displays "1. 2." (i.e., plays the volta twice) the volta end repeat measure's 'play count', should be set to 3 -- to send playback to the start repeat barline twice (thereby changing the playback count from 1 to 3.

https://musescore.org/en/handbook/3/repeats-and-jumps#adjust_repeat_pla…
and:
https://musescore.org/en/handbook/3/measure-operations#play-count
and:
https://musescore.org/en/handbook/3/voltas#Play_count
which shows a more complex volta.

Reply

• Dec 11, 2021 - 21:54

My point was that why the measure (bar) has repeat count but not the repeat-barline (dotted double barline) itself. And, the measure (bar) repeat count does not show in the inspector. It only shows when one looks into the "bar property". To me, it is more intuitive to see the repeat number in the inspector and when the repeat barline is selected. Particularly given that repeat barline is often used without e.g. "1.,2.,3.,4.". However, I think that I understand that use of bar repeat count property is probably to accomodate both the use of "1.,2.,3.,4." and the use of repeat-barline only.

Reply

• Dec 11, 2021 - 23:41

Thanks for that. I understand that it is to do with design decisions, made from time to time to figure out what the best UI. But the first priority was, of course, to provide all the functionalities needed. Some of the functionalities may not be in the best places - not intuitive enough (yet?).

Reply

• Dec 12, 2021 - 02:18

You wrote:
As to your revised score, it applies Fine only at the 5th repeat... To me, however, that is rather strange. That is not intuitive. It means, D.S. al Fine applies only at the 4ht repeat and Fine applies only at the 5th repeat.

Actually, because of the 4 verses, there exist just 4 repeats. The Fine is applied after a final chorus playthrough, not a 5th repeat.
Here's why:
A jump (D.C., D.S., etc.) which appears under a volta is taken on the last normal pass of the measure containing it. So in my score, after the 4th verse plays, the jump (D.S. al Fine) is taken. You consider it a 5th repeat, but technically it is a jump (which could have landed at some other place instead of the chorus). It is this jump, not a repeat, which just happens to cause a 5th chorus playthrough -- and so not listed in the Inspector's 'repeat list' as a number '5'.

Your observation of the measure play count being in Measure (Bar) Properties rather than a property of the repeat barline itself has come up in the past. As Jojo mentions, it's been that way for some time.
MuseScorer jeetee is the current Repeat, Jump, Volta, Play Count guru and maintains an impressive collection of tests here:
https://github.com/musescore/MuseScore/tree/3.x/mtest/libmscore/repeat
(Some really complex scenarios)

Concerning design decisions and underlying logic, I leave that up to the actual developers, and rely on the handbook and these forums for providing enlightenment.
Still, I'm in awe when I try something like this:
Crazy_Happy_Birthday.mscz
and it works!

Reply

• Dec 12, 2021 - 02:43

I understand what you are on about, but when D.S. al Fine (Jump) should apply is not intuitive when it is combined with 1,2,3,4 (and a dotted repeat double-barline in this case), for those who do not have any idea that the 1,2,3,4 should take precedence over D.S. al Fine "Jump" only after "play though". To me, this seems like a particular problem with digital score with auto-play function. With printed scores, D.S. al Fine is written BEFORE the repeat barline and it looks as though it should be applied even for the first repeat. So, if Fine is encountered during the first repeat (the measure play count is actually 2), the song should end there, invalidating any further progression. (Though, however, this is not the case with Play with MuseScore as you demonstrated.) For this reason, it is clearer to use the 5th Volta in the last measure of the Chorus (one added following note tied with the proceeding "cry" note). This is easier to understand and that's how the original score is written. It is clear indicative that that is preferred. In order for this transcript to be easily comparable, keeping the original structure is preferable.

Reply

• Dec 12, 2021 - 04:48

Earlier you wrote:
The original printed score did not use 1,2,3,4 nor 5. but I thought they make it easier to understand.

So I thought you were adding the voltas.
But I may be misunderstanding something because you most recently wrote:
This is easier to understand and that's how the original score is written is clear indicative.

In any event, you managed to get the score to your personal liking. I guess that's what really matters.

You wrote:
...when D.S. al Fine (Jump) should apply is not intuitive when it is combined with 1,2,3,4

Jumps under a volta (e.g., 1.2.3.4.) are taken on the last pass.

You wrote:
With printed scores, what one "reads" from the score like mine is that when 1,2,3,4 and the repeat barline instructs the repeats, D.S. al Fine should apply as well - every time. So, if Fine were reached along the way, the song should end there, invalidating any further progression.

Again, jumps (D.S. al Fine) under a volta are taken on the last pass.

Also, if Fine were reached along the way, the song should end there, invalidating any further progression.
Really?
There are millions of songs which have a D.C. (or D.S.) al Fine at the end and where the Fine is somewhere in the middle.
For example: Skye Boat Song_3.mscz
Are you saying that you would stop at the 'Fine' and never play the rest of the song?
Most everyone else would make a mental note of where the 'Fine' is, play until the D.C. al Fine, return to the beginning and then stop when the 'Fine' is reached.

Reply

• Dec 12, 2021 - 05:15

Thanks for that. I was lacking in that knowledge. In my little experience, usually Fine is only used in combination with D.C. al Fine or D.S. al Fine, not together with Voltas. As you demonstrated (and it must be common knowledge), obviously, when D.C. or D.S. are used within Voltas, they are applied only after the Voltas. Since the original score does not use Volta, it only leaves a text instruction "last time only" above the extra measure concerned. So, I only added the 5th Volta to that instruction. The reason is obvious. Otherwise, the score does not play properly with MuseScore.

Reply

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