downloaded score plays differntly than direct playback of score from library search
I was having an issue with repeats and playback so I downloaded a version of Blue Danube Op314 (r_stonge upload). I could not get the downloaded score to playback as I thought it should. I went back to the Musescore library search and played the score directly in the browser. Starting at measure 95 which is marked as the beginning of the section named No. 2. It played exactly as I thought it should. I won't go through the details but all the skipped measures and jumps and playbacks were how I was trying to get the local program to function but on my computer, the downloaded score could not replicate what I observed in the browser playback. If someone wants details I will include them in a reply. I do not have a score to upload as it was the direct download with no modifications.
Comments
In order to understand and assist, we'd definitely need more information - precse steps to reproduce the problem. And if you don't wish to upload the score, we'd also need a direct link to it (realize, though, it may have been edited since you downloaded it, so there is no guarantee we'll be accessing the same score as you).
As it is, my guess is that you simply have the repeats returned off and need to re-enable them by clicking the button on the toolbar. Either that or the score in question was uploaded using a different version that handled repeats differently.
In reply to In order to understand and… by Marc Sabatella
I tried to upload the file which I downloaded and have made not changes. It is exactly as received. However the system says it exceeds 4 MB and won't upload. All I did to play the downloaded file is to keep current style, leaving whatever was downloaded in place. Repeats are enabled on toolbar and are taken but the path is different in local vs the online playback
The section in question has M95 as a lead in. M96 starts a repeat and is segno. The barline after M111 ends repeat. Volta 1 spans M111. Volta 2 spans M112. Volta 3 spans M113. There is no second end repeat between 112 and 113. Playcount is 2 and volta playlists are 1,2, and 3 respectively. M114 continues on to M129 which ends the section and is a DS al Fine. M113 (with Volta 3) contains the Fine.
Online playback is M95 to m96 through M111 and repeat returns to M96 playing to M110 and skipping M111 (volta 1) but plays M112 (volta 2.) M113 (volta 3) does not play and continues M114 through M129 which jumps back to segno M96 and plays the repeat through M110 (before any voltas.) M111 and M112 (voltas 1 and 2) are skipped and M113 (volta 3) plays and as this is fine, the score jumps to next section M130.
In the downloaded file, the first repeats are as above. However the DS al Fine pass plays all the repeats again just like the first time and never gets to the Volta 3 with the fine and so after starting at M114 finishes the section at M129 and continues. The DS has Play Repeats specified. If I uncheck play repeats, the DS pass plays up to the voltas and then only plays Volta 2, the continues at M114 and finishes out. Neither option plays the way the online version plays and supposedly it is the same file being played. The issue is the way the online system plays back is the way I believe the score is intended to be played. (Voltas 1 and 2 on first pass, and only Volta 3 on DS pass.) So what piece of software plays the online version? When the file is opened it says it was created in 3.4.2.
In reply to I tried to upload the file… by msokol
3.5.1 came with a bunch of bug fixes for volta playback and jumps.
From your description though (again, much easier to follow along if you share the score url here) it seems as there indeed is no real 3rd repeat in the score at all, so no way for the jump to know that it exists and thus take the 3rd volta after that.
You can force MuseScore to understand the higher number of total repeats by increasing the playcount for m111 by one. The volta over it will prevent the repeat from actually being taken more than once, but the increased count will be taken into account when determining what the "last repeat" is that should then be used after the jump.
In reply to 3.5 came with a bunch of bug… by jeetee
I will try that. I also have the option you provided several months ago (since modified for my needs) to move the third volta measure to the end of the section and use jumps to navigate. In this situation, I replace the DS al Fine at the end of the section with a DS al Coda. I add the coda to the moved measure and put a to coda jump at the measure before all the voltas. I am attaching my sample file.
In reply to 3.5 came with a bunch of bug… by jeetee
Set play count to 3 and tried it and it did not work. The DS al Fine jump doesn't occur and score moves on to next section. Is this perhaps because the play count is not 3 and thus the jump does not occur because it is not equal to the highest play count from the play lists?
In reply to Set play count to 3 and… by msokol
Really.. the reason for the original post was why does the online player play differently. Is it not using Musescore 3.6.2?
In reply to Really.. the reason for the… by msokol
The backend on musescore.com is using MuseScore 3.6.1
In reply to It is 3.6.1 by Jojo-Schmitz
Well.. for new uploads it is.
For other scores is is playing whatever it was using back when those were uploaded.
In reply to Well.. for new uploads it is… by jeetee
Which is why if you listen to scores from 10 years ago, you can still hear them playing with the original MuseScore 1 soundfont!
In reply to 3.5 came with a bunch of bug… by jeetee
An alternative to moving the third volta measure to the end and using jumps is to use the scheme in the attached. In this version, which I posted a few months ago, the solution is to interchange voltas 2 and 3 and add the second end repeat after what is know volta 3 (in the second position.) This works .. plays voltas 1 and 2 on first pass and only the volta 3 on the DS al fine pass. The only issue I have is that in the Blue Danube score I have referenced, inserting this option again fails to play the DS jump and exits the section. This is the same problem I mentioned on JeeTee's comment the other day that the DS pass was not being taken either. something in the score is not right. In the Danube score, need to make sure play repeats is unchecked in DS command as original had it checked as well as the reordering I noted.
In reply to An alternative to moving the… by msokol
I discovered that to make the DS al Fine jump to work it must be part of a repeat. In my MixedUpvoltas2 score, I had a second repeat at the end of the section because the score I used for my prototype had such a structure. The DS is in Volta 2 of this second repeat. In the Blue Danube, there is no second repeat, just simple measures and in this construct the DS jump is not taken. If however I make the last measure of the section a single repeat with play count of 1, the jump is taken. Just put begin and end repeats around the measure with no voltas and changed play count. I speculate that something with the first repeat not reaching a value of 3 has something to do with the non-jump but whatever is being checked by the system is reset by a second repeat of some fashion.
In reply to I discovered that to make… by msokol
I don't think your assessment is entirely spot on and a jump instruction does not have to be in a repeat measure at all to be honored. And although I'd keep the 2nd volta in the end measure of this example, you can see that it is not required at all (and the repeat barline at the Fine is just confusing, better use a real end barline instead)
In reply to I don't think your… by jeetee
I understand your comment. The reason for the repeat at the very end has nothing to do with my schematic where the DS jump is taken. When I applied the mixed up triple volta to the acutual Blue Banube score in mentioned previously ( where I reordered the voltas in order 1,3,2 ) and did nothing else for the remainder of the section, the jump is never taken and goes to next section. In the Blue Danube there are no other repeats after the initial repeat. I found that the only way to get the jump to be taken was to create the single measure repeat. When I do this in the Blue Danube score, the dump is taken and only plays the second volta which is Fine and exits. The reason for this situation in the Blue Danube score confuses me. I will try to extract only this section to a stand alone file and see if the behavior remains. If it does, I will upload it.
In reply to I understand your comment. … by msokol
If you could share the link to the full score, then we could inspect it as well.
In reply to If you could share the link… by jeetee
This is the link of the score from a search of libraries on Muse Score. This is the one I started with.
https://musescore.com/r_stonge/the-blue-danube-op-314-piano-johann-stra…
There is something about this file and the download file is over 4MB and even if I cut most of the score and resave the file does not shrink so I cannot upload the extract I made with just the section (No. 2. ) in question. Essentially I just reordered the measures in the first repeat, containing the fine, and used the construction in the previous thread of comments. There is not other repeat in this section and ends with the DS al Fine.
In reply to This is the link of the… by msokol
The file size was due to the insanely high quality cover page; see the attached in which I've reduced that. I'll look into the repeat issue later; but here are some immediate giveaways:
A./ I would expect section breaks between each "No x" (well... section) of the score, there aren't any.
B./ For some reason the last coda system has been made as a picture instead of written in the score.
In reply to The file size was due to the… by jeetee
When you increase the play count for the 1st volta measure, to allow it to use a 3rd volta after the jump, you also say that the 3rd repeat is the "last" normal repeat to consider. So when you then reach the D.S. during the 2nd repeat, it is being ignored as a Jump should only be honored during the "final" repeat for the measure containing it..
So you need a way of telling MuseScore that the "final" playback for the measure with the D.S. is the 2nd repeat; for example by covering it with a(n invisible) 2nd volta.
In reply to When you increase the play… by jeetee
Wow... not exactly intuitive.... The final section after the first repeat until the very end is simple measures without any repeats so why is not anytime through it a final repeat? And adding the hidden volta when there are no repeats around is used to create a situation that corrects for the final play count being 3 in another repeat? So if I understand, and I am really fuzzy on the algorithm, the first repeat never gets to a play count of 3, because it cannot. Then when we get to the final measure with the jump, the system remembers the play count from a measure that does not contain the jump and says I am not at that number yet. So adding the hidden volta of 2 results in a reanalysis of the situation and the system only sees the volta and the play list value of 2 is the highest value with that "repeat" (which does not exist so this is a rouge volta?) and says okay you match the current play count, it must be final repeat so take the jump. The takeaway from this is that the status of a repeat at the beginning of the section (as to matching the highest play count possible) carries throughout the balance of the section and there is no actual repeat associated with location of the jump.
I appreciate the time and effort. Not sure what you did to reduce the graphics. I tried deleting everything but section 2 and still the size did not reduce so there must have been something hidden that I could not see.
In reply to Wow... not exactly intuitive… by msokol
Here is the important part you might not get:
Repeats are always relative to their starting point reference
That can be either the score start, or a section start, or a start repeat barline.
To figure out what the final repeat is, you have to regard all play counts from all end repeat barline( measure)s.
Upon a Jump, it doesn't remember the play count from another measure, but starts looking backwards for a start-repeat reference point from the jump target and then uses the "final" count evaluated for it.
So when MuseScore has to evaluate whether a Jump must be taken, it indeed compares the current play count (again, relative to the starting repeat point) with the final repeat count for that same reference point; If they match, then we are doing the "final" playback and thus should honor the Jump.
Now volta's change this behavior in that their "repeatList" is also taken into consideration and that when the current play count matches the highest number in the repeatList; it is also regarded as the "final" playback.
What the hidden volta thus does is trigger that additional volta repeatList logic evaluation, by matching its repeatList value with the actual current playback count of that measure. (make no mistake, the repeat does exist and is still relative to that same starting repeat barline).
> "The takeaway from this is that the status of a repeat at the beginning of the section (as to matching the highest play count possible) carries throughout the balance of the section and there is no actual repeat associated with location of the jump."
So no, that is not the takeway; because there is a repeat associated with the location of the jump in that score.
The real takeway should be "if you deviate from standard notation and then abuse voltas for what usually is a coda and thus have to fake the repeat count to be one more than under normal circumstances, then you might also have to fake the repeat/jump evaluation logic to compensate for the earlier abuse of such a feature."
.. which is also why it is less intuitive; as it is a hack.
As for reducing the image size; even after removing an image from the score, it is still maintained within the mscz-archive due to a bug. I manually removed it from the archive, then resized it and added it back in.
In reply to Here is the important part… by jeetee
Thanks for the explanation again.. so in the absence of the hidden volta the play count is relative to the only start repeat line in the section which would require a play count of three to be considered a final playback. As to nonstandard notation, it is not my scoring but that of Johann Strauss and by the way, it appears to be a general method of scoring in similar waltz music of the era.
In reply to This is the link of the… by msokol
"UPLOADED ON MAY 10, 2019" so it probably doesn't have the latest repeats fixes? That might explain why it plays back differently on musescore.com vs. in the lates version of MuseScore. Back then 3.0.5 was current and as such used to render the score on musescore.com