End of Life plan for 3.x ?
I've been using 3.6.2 and it works fine for my simple needs; I'd rather not upgrade to 4.x.
I did some searching, but can't find anything mentioning how long support ( mostly interested in security fixes ) is planned for 3.x
Is there a support plan for 3.x?
thx
-tom
Comments
In reply to I don't think that there… by yonah_ag
But they are still compatible? Everything possible in 3.6 is possible in 3.7, this included
In reply to I don't think that there… by yonah_ag
A 100% layout compatibility definitively means that this header-issue should be reverted...
In reply to Yes, so a 100% layout… by HeinzVo
I'm not sure what this issue is you are referring to, is something not possible in 3.7 that you can do in 3.6?
In reply to I'm not sure what this issue… by XiaoMigros
No, but the 'bugfix' will mean that 3.6 and 3.7 lay out headers differently. I don't want to have to revise my headers or run 3.6.2 for my older scores.
In reply to No, but the 'bugfix' will… by yonah_ag
If the decision is to retain the bug fix in 3.7, one possibility might be to write a script to make the changes needed automatically in the MSCX files. I don't know how difficult this might be, but I've done something similar for batch conversion of scores to bass clef.
In reply to If the decision is to retain… by BanjoJake
Yes, but it's anyway not a good decision to leave or retain bugs in an app. And Marc classified that behaviour with that incorrect space calculation as bug as well...
In reply to Yes, but it's anyway not a… by HeinzVo
exactly, its not a good decision to leave bugs in an app. so lets let 3.7 fix the issue from 3.6.2
In reply to exactly, its not a good… by XiaoMigros
[oops]
In reply to exactly, its not a good… by XiaoMigros
Normally I would agree but here scores shared online could have a broken layout for users who download to 3.6.2. What will the server side do with the differences?
In reply to Normally I would agree but… by yonah_ag
that is already the case for mu4 scores in mu3.6, so there is little value in this argument imo
In reply to that is already the case for… by XiaoMigros
A poor analogy. It is quite reasonable, and fully expected, that MS4 scores will be incompatible with MS3.6.2 but I think that a lot of consideration should be made before making 3.7 scores deviate from full compatibility with 3.6.2 unless they are prevented from being uploaded to musescore.com.
Where does this end? The more that 3.7 diverges in compatibility with 3.6.2, the less attractive it becomes as an update. Maybe the severity of the bug merits this deviation, but maybe not.
In reply to A poor analogy. It is quite… by yonah_ag
There is still a considerable number of users of MS3.6.2, not only because of security or privacy concerns related to Muse Hub but also because as yet it is far from allowing all the functionality of MS3.6.2 and it is more hardware demanding
I think there should be sort of a compatibility plugin so that a MS4 score could be opened with MS3.6.2 keeping at least most of its layout. The workaround is to export from MS4.2 to .musicxml or .mxl and open it from MS3.6.2, but the layout is ruined.
Another option would be that MS4.2 could also export to a MS3-compatible .mscz format
In reply to There is still a… by fmiyara
There's still my 3.7...
In reply to There's still my 3.7... by Jojo-Schmitz
I was referring to the last official version.
By the way, so far I couldn't succeed locating a downloadable version of 3.7. It's my fault, I know, but the links you have provided lead to directories with a structure I don't understand. I'm not a developer and I ignore even the most basic terminology regarding github. I find a lot of folders and files with unfamiliar extensions and I am completely at a loss with all that stuff. I would love to try MS3.7 if only I could locate and download it.
In reply to I was referring to the last… by fmiyara
Download and installation instructions can be found here:
https://sites.google.com/view/musescore-3-evolution/
In reply to If the decision is to retain… by BanjoJake
Re: BanjoJake • Aug 12, 2023 - 01:10
If the decision is to retain the bug fix in 3.7
If I fix my scores with such a script then they won't display correctly for people using 3.6.2, so they would no longer be suitable for sharing.
In reply to If the decision is to retain… by BanjoJake
Something like that script would be needed in master/4.x too (and first)
In reply to Indeed, and there is… by Marc Sabatella
This there is actually a bug here worth fixing: the algorithm that computes the amount of space required for the header allocates (way) too much space if you use the old-style hack of applying a negative offset to force the header into the margin should get mentioned in the issue (https://github.com/musescore/MuseScore/issues/19014)
Edit: it is...
In reply to This there is actually a bug… by Jojo-Schmitz
What do you think about compatibility with 3.6.2, (including downloads and server side rendering)? As the main developer you are probably best placed to make this decision.
In reply to What do you think about… by yonah_ag
If and when it get fixed in Mu4, I'll backport it.
Or if and when someone creates a PR against my repo's 3.x branch, I'll seriously consider it.
In reply to This there is actually a bug… by Jojo-Schmitz
NOTE: It calculates as well not needed extra space even at positive offsets. So it calculates in any case too much space!
In reply to Reg. 1.: I might have nailed… by Jojo-Schmitz
YES, it worked ! Plugin Manager is populated, even after clicking 'Reload'. The ABC ImpEx plug-in worked. MuseScore version (64-bit): 3.7.0.5784330917, revision: 0b546f0 on Windows 10 Version 2009 or later, Arch.: x86_64. THANK YOU !!!
In reply to YES, it worked ! Plugin… by BanjoJake
Thanks for testing and confirming
In reply to Reg. 1.: I might have nailed… by Jojo-Schmitz
Hello,
It seem to work perfect here.
With the latest windows build the header-foorter is fine also I suppose.
One small problem.
I have 14" laptop with 150% scaling in order to be able to read the letters.
Musescore has some problem taking into account that scaling.
I suppose it has to do with Qt (dorico had also that scaling problem in earlier versions)
I can fix it with a compatibility HDPI setting in windows for musescore.exe
Maybe there is a better solution for that.
Thank you and kind regards,
In reply to Hello, It seem to work… by Johan-v
I don't have (or rather don't know) a better solution for this.
In reply to Hello, It seem to work… by Johan-v
Yes, I recognized that as well (but also in MS3.6.2). A scaling of 100..149% works well, but >=150% really causes some problems - especially in displaying the mixer..
In reply to Yes, I recognized that as… by HeinzVo
I do not have that problem in 3.6
In reply to I do not have that problem… by Johan-v
What about the 32-bit version of 3.7?
In reply to What about the 32-bit… by Jojo-Schmitz
The (mis)scaling problem at >=150% windows-screen-scaling (especially the much too big mixer sliders) occures in all versions MS3.6.2 to 3.7.0, 32-bit and 64-bit. The sliders go to just about half the scale, if switched from 150% to 149% or lower of windows-display-scaling.
In reply to What about the 32-bit… by Jojo-Schmitz
I don't have the problem with the 32 bit version.
So I suppose my scaling difference has to do with the used Qt version.
In reply to I don't have the problem… by Johan-v
That's indeed my suspicion
In reply to That's indeed my suspicion by Jojo-Schmitz
FWIW, I have the same "issue" (I prefer the scaling of 3.7) with 3.6.3 through 3.7, as well as any version of MuseScore 3 portable. Does the portableapp use a different version of Qt as well?
For Windows - do I need to compile 3.7 or is there a pre-complied package available ?
In reply to For Windows - do I need to… by rothers
Precompiled versions are available. See https://github.com/Jojo-Schmitz/MuseScore/actions and drill down to the artifact
In reply to Precompiled version ate… by Jojo-Schmitz
Thanks, so is this the latest ?
https://github.com/Jojo-Schmitz/MuseScore/actions/runs/5544645636
In reply to Thanks, so is this the… by rothers
Look for the lastest for the 3.x branch
In reply to Look for the lastest for the… by Jojo-Schmitz
I would like to give some suggestions :
1 - Have a webpage for the 3.7 with a download link.
2 - Backport the double-click to enter edit mode feature.
3 - Improve zerbeus volume. Some instruments that are correctly leveled at Sforzando and at Sfizz get audible volume differences in Zerbeus, when switching from one dynamic layer to another or among different instruments at drumkits. Sfizz works much like Sforzando and if zerbeus did the same it would provide nice support for sfz.
In reply to I would like to give some… by fernandoamartin
In reply to I would like to give some… by fernandoamartin
1 - A Musescore 3.7 webpage (musescore-classic.org?) would be nice.
In reply to A Musescore 3.7 webpage … by yonah_ag
Sure. I could do that, if I'd find a sponsor ;-)
In reply to Sure. I could do that, if I… by Jojo-Schmitz
My 2'cents Proposal:
1) https://github.com/Jojo-Schmitz/MuseScore/ is a fork of Musescore/Musescore.
As a fork repository, contribution doesn't count towards your activity in your (or other contributors github profile), this could discourage contribution. By copying the repo into a fresh github repo the fork link is removed. The upside of this is that 3.x branch (which as of now is kind of esoteric since its upstream doesn't exist) will be the new 'main' branch.
2) New code/contributions is formatted as Musescore4, yes this will result in mixed formatting. I'm doing it currently and find no problems reading the old format, or mixed formatted code. It has the upside of easy seeing what is older and should be handled more gently (yes, git blames does that also).
In reply to My 2'cents Proposal: 1)… by larryz
1) I don't understand this. But even in the upstream preo only non-merge commits to master counted as contributions. I'M only workong on the 3.x branch in my repo, under the assumption that this won't change anymore upstream.
2) On my 3.x branch I'll maintain the 3.x style. That BTW is one of the problems of your album PRs, albeit the least ot them.
In reply to 1) I don't understand this… by Jojo-Schmitz
Yes, lets keep 3x as a maintenance branch to support 3.7 being in maintenance mode.
And lets do the more fun stuff in 3.8 ;)
In reply to Sure. I could do that, if I… by Jojo-Schmitz
Re: Jojo-Schmitz • Aug 11, 2023 - 09:17
Sure. I could do that, if I'd find a sponsor ;-)
Are you on Patreon?
In reply to Are you on Patreon? by yonah_ag
no
In reply to Precompiled version ate… by Jojo-Schmitz
I just want to add my appreciation for all the work you've done on 3.7. Hope we could have some kind of "official" release in future. Cheers.
In reply to I just want to add my… by Nuno R. Rodrigues
Not very likely...
I've just managed to get 3.7 to open Mu4 MSCZ files.
In reply to Not very likely... I've just… by Jojo-Schmitz
Great. If you ever setup a funding/sponsoring channel I'm sure that lots of non-developer supporters would contribute (I would). I would also hazard that a fork would be a project much more appealing for open-source developers than the direction where "official" MuseScore is going.
In reply to Great. If you ever setup a… by Nuno R. Rodrigues
+1 for funding/sponsoring
In reply to +1 for sponsoring by yonah_ag
I would contribute toward funding 3.7.
Thanks SOOO much for the 4.x to 3.7 import!!
In reply to Not very likely... I've just… by Jojo-Schmitz
Thank you ! It still gives the error message, but after clicking "Ignore", then seems to work (maybe with some formatting quirks?)
In reply to Thank you ! It still… by BanjoJake
Yes, that Ignore is needed (and it not an error but a warning) and yes, all styles are set to their defaults, that's a problem I still have to solve, would need to read in the style file from the Mu4 .mscz too. (and not only the style file, like without the code change)
In reply to Yes, that Ignore is needed … by Jojo-Schmitz
(not a big deal, but the the pop-up box header actually says "Load Error" - in Windows)
Anyway, I'm able to attach the files to my v3.6.2 style (.mss file), so that's pretty great ! Thanks again !!
In reply to (not a big deal, but the the… by BanjoJake
Well, no, as per the code, and the icon in the dialog, it is a warning. Despite the wording "Load error"
In reply to Yes, that Ignore is needed … by Jojo-Schmitz
One step further: it now also loads pictures, should the score have any.
In reply to Yes, that Ignore is needed … by Jojo-Schmitz
And now style too!
In reply to And now style too! by Jojo-Schmitz
Fantastic!
In reply to Yes, that Ignore is needed … by Jojo-Schmitz
And now it also restores the view mode.
Tempo notes are smaller in MS3.7
:
This is MS3.6.2
and here is exactly the same score, unchanged, in 3.7.0
As you can see, the tempo crotchet symbol is sized differently even though both are Edwin 12 pt.
In reply to tempo notes are smaller in… by yonah_ag
I believe that is because it doesn't take that qlyph from Edwin, but from the Musical (Text) Font, Leland, and that had been updated in 3.7 (and master, 4.x), I guess Leland now has that glyph and before didn't have it, so fell back to using Bravura
In reply to tempo notes are smaller in… by yonah_ag
Is there any workaround for this ? Can the note size be corrected in inspector ?
In reply to Is there any workaround for… by rothers
Just found it, change the note font to Leland in text edit mode.
In reply to Just found it, change the… by rothers
I was able to do this once, but now the "note font" entry in text edit mode seems to have disappeared, or been moved. I'm running MuseScore-3.7.0.7201067451-x86_64 on Linux Mint, if that's useful.
In reply to Is there any workaround for… by rothers
Yes, of course, the issue is only present when no changes have been made
In reply to tempo notes are smaller in… by yonah_ag
Would it be possible to post workarounds like this on the Web page? I just searched this entire topic thread to find this one again.
In reply to Would it be possible to post… by dhfx
On what webpage?
In reply to On what webpage? by Jojo-Schmitz
https://sites.google.com/view/musescore-3-evolution/
In reply to https://sites.google.com… by dhfx
Ah, that'd be something for @yonah_ag then
In reply to https://sites.google.com… by dhfx
I'll try to collect some of these and make a subpage for them but it could be a while before I find some spare time.
Congratulations and thanks Jojo. Now the backward compatibility of MU3 openning MU4 files is working nicely. We can finally switch back and forth between the freedom of choice of 3.7 and the realism of Muse sounds in 4.x.
In reply to Congratulations and thanks… by fernandoamartin
This might grow to a very interesting feature. Many thanks for that!
But at that very first implementation state some really important things go lost during import (eg crescendo lines, expression texts and so on or also some text styles like "Dynamics"). All other elements they do import (and are not auto-placed) unfortunately are not scaled by the "Format / Page Settings / Staff Space" factor during import, so they all will not lie at the initially defined correct positions in the score and must be moved and/or resized after importing (see https://github.com/Jojo-Schmitz/MuseScore/issues/165). This also includes element positions, distance spacers and many other defined distances (eg gap settings before and after frames or images).
We hope this could be corrected as soon as possible, so that interesting import functionality might get practical and useful in future time.
In reply to This might grow to a very… by HeinzVo
Please report those too, one element per report
In reply to Please report those too, one… by Jojo-Schmitz
Thanks Jojos very hard work during the last days/weeks the backimport of MS4-scores to MS3.7 (MS3 Evolution) now works really fine. That was a great step or improvement - many thanks to you.
In reply to Thanks Jojos very hard work… by HeinzVo
Thanks. I'm still on it though .-)
In reply to Thanks Jojos very hard work… by HeinzVo
I believe Mu4 import in my Mu3.7 is pretty decent now.
A Mu3 -> Mu4 -> Mu3 roundtrip still does loose a few things:
All these because they get lost on import of Mu3 scores in Mu4 already, so nothing I can do about those, other than reporting that as bug (for those cases that are not by design, see also cmd.cpp,
resetDefaults(), lines 2463-2525), which has been done.It also looses stuff that exists only in Mu4, like 2- and 4-measure repeats, straight flags, etc., nothing can get done about that other than backporting those features, which is way too big a task.
It also looses parts, so those need to get generated (and formatted) again, I might eventually be able to fix that. Or maybe not...
If you find more issue with this: just let me know, preferably via an issue in my fork.
In reply to I believe Mu4 import in my… by Jojo-Schmitz
Yes, I believe the "Play/NotPlay-Flag" for Trills get lost too.
In reply to Yes, I believe the "Play… by HeinzVo
Yes, added to the above list and reported on GitHub
In reply to Thanks Jojos very hard work… by HeinzVo
And some special thanks to @HeinzVo for reporting and testing ;-)
In reply to Congratulations and thanks… by fernandoamartin
Thank you for making this 3.7 fork.
I use MS3 because it is like a Swiss Knife for classroom and education.
I understand what MS3 and MS4 are missing compared to the big boys. (sibelius, dorico, finale, online apps)
MS4 is trying to fill that gap but at the cost of its biggest strength for my use.
Maybe MS4 will become again that swiss army knife but for now it is not.
So I'm glad 3.6 continues to live with this 3.7 fork.
Thanks.
In reply to Thank you for making this 3… by Johan-v
Some of your reasons seem flawed:
Mu4 is available in different languages too, I believe it has the same set of audio drivers as Mu3, it works the same on all computers too, as long as you don't use MuseSound the file size is comparable to Mu3, Mu3 does have its share of audio problems too
In reply to Some of your reasons seem… by Jojo-Schmitz
I knew this was comming.
I can only share my experience.
In reply to I knew this was comming. I… by Johan-v
That's fine.
In reply to Thank you for making this 3… by Johan-v
The unavailability of Mu4 as a PortableApp is at least known and being worked at.
In reply to The unavailability of Mu4 as… by Jojo-Schmitz
It is already available on portable apps but not here.
In reply to It is already available on… by Johan-v
I know... and Casper is working on it, so it'll hopefully be there for 4.2. I don't think it'll get retrofitted for 4.1.1
Is there a way to use the new MuseSounds in 3.7.x for Linux? I'd like especially to be able to use the ensemble string sounds, instead of using one violin/viola/cello/bass to represent a whole section.
In reply to Is there a way to use the… by dhfx
No
In reply to No by Jojo-Schmitz
"No", as in: "currently not possible", or as in "will never be possible" (and if so, for what reason?)
In reply to "No", as in: "currently not… by johnweigand
Mu4's new sound system is way too complex to get backported
In reply to Mu4's new sound system is… by Jojo-Schmitz
Thanks for answering, I missed it for some reason at the time.
I believe you, however I would still like to understand what's making it so complex. The code looks daunting to me. Do you know of any documentation on it?
In reply to Is there a way to use the… by dhfx
You can choose "violins" instead of "violin" when selecting the instruments. While not perfect, it sounds decent, and certainly more similar to a violin section than to a single violin. The same hods for the other strings.
In reply to Is there a way to use the… by dhfx
You may be able to use MIDI Out or JACK with MS3 on Linux to route audio to VST3 for playback. It certainly works well with Windows and seems to be supported on Linux.
https://askubuntu.com/questions/526275/is-there-any-virtual-midi-loopba…
Thank you @Jojo-Schmidt for trying to reward loyal MS3 users who dont want to change! I was so happy to use MS4 till i found it it doesnt work with windows 7. So i figured i'll be with 3.6.2 till i die.
I was happy to see youre trying to bring in some of the features of 4 back into 3, and saw youll be including VST support, so with that i decided to start a useless GitHub acct just so i can have that one feature. (I also wanted a better piano roll editor, but I'll work with the clunky old one if I can just use my orchestral VSTi's.)
I have successfully followed the instructions you gave on the dedicated web page for Evolution, and it opens fine and creates a new score on my Windows 7 64-bit computer. I am unable however to load VSTi's. When I open the synthesizer menu I'm only seeing options to add sound font files. I went into preferences to see if there was a special folder path that needs to be set, but that wasn't there. Is there a special trick to be able to load the plug-in instruments?
And also where can I find a list of changes you've made to test your added features?
In reply to Thank you @Jojo-Schmidt for… by harbinger
One way is to use the IO tab of preferences to send MIDI Out to a VST3. This also works for sending the output to a DAW.
See https://musescore.org/en/node/358470
and https://musescore.org/en/node/339943
In reply to Thank you @Jojo-Schmidt for… by harbinger
There has been a recent update to the Piano Roll Editor in 3.7
https://musescore.org/en/node/359866
In reply to Thank you @Jojo-Schmidt for… by harbinger
Jojo-Schmitz please ;-)
Wanting to stay on Windows 7 for eternity is the worst reasin to be using Mu3...
Mu3 doesn't work with VTSi, not direktly anyhow, not in the way Mu4 does
Pianorol editor just got some updates
The changes are in the Git history
In reply to Thank you @Jojo-Schmidt for… by harbinger
Never say you are too old. My uncle was 101 years old when he decided to buy a new computer because his old one was to slow.
I've now copied @yonah_ag's https://sites.google.com/view/musescore-3-evolution/ to https://github.com/Jojo-Schmitz/MuseScore/wiki/MuseScore-3-Evolution
As of today 3.7 also fully imports all style settings from Mu4 score (all that exist in Mu3 too or have a decent counterpart), special cudos to @workdwideweary to spot the bad bug of mine that prevented it from happening before.
So how about for Musescore 3.8 Evolution - same as 3.7 but with the engraving improvements of 4.* (but not the interface or musesounds of 4.*).
In reply to So how about for Musescore 3… by rothers
Be my guest ;-)
Some of the engraving improvements have been ported, some are way to complex
In reply to So how about for Musescore 3… by rothers
Isn't every MS3 version after 3.6.2 still 3.7 Evolution?
In reply to Isn't every MS3 version… by yonah_ag
Yes
In reply to Yes by Jojo-Schmitz
And that post (of yours) makes 500 in this thread! Clearly a popular topic. MS3 is alive and well.
I have been running 3.6x for ages (MacOS) and was just trying to install 3.7, followed instructions on the site. I downloaded the most recent version, it seemed to act like a normal .dmg file, but after installing it, the computer says that the application file is damaged and should be moved to the trash. I am not a computer expert, just a user, and I am confused... what should I be doing?
In reply to I have been running 3.6x for… by ayermish
@ayermish
There are challenges when installing and running on a Mac.
First, the application is not digitally ''signed" so you have to convince the Mac that you really want to run it. Here's a link to a discussion that includes installation instructions.
Secondly, somewhere after ** macOS version MuseScore 3.7.0.4524440406, revision: f3d36a3** MuseScore 3.7 no longer sees any audio devices ... meaning it can't playback any sound.
You can get the version I mentioned, but:
• Surely there's a later version that still has functioning audio, but I don't know what that is.
• Hopefully development will solve the audio problem so those using MS3.7 can use the latest release
scorster
In reply to Sounds like you're on a Mac? by scorster
The one from here https://github.com/Jojo-Schmitz/MuseScore/actions/runs/10717483411 is newer, from a week ago.
Assuming the audio issue is related to the Qt version used to build
If you need a newer version drop me a note in https://github.com/Jojo-Schmitz/MuseScore/pull/93