End of Life plan for 3.x ?

• tammi 13, 2023 - 03:15

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


Kommentit

In reply to by Jojo-Schmitz

Yeah, you definitely shouldn't stop support for 3.x.
1. It works on older machines.
2. It is fine for most 'old-school' composers, who don't use the computer as an aid/crutch to composing but us the computer as a pure notation tool to write already composed music.
3. It uses, what I suspect is the 'old code' that may be disorganized, but has 7 - 10 years of updates from users and programmers embedded into it.
4. It cannot be integrated with Musehub and is the last stand alone pure compositional aid.

I hope you can become a power to be and override this decision.

In reply to by fernandoamartin

I think it's a good way.
If Jojo(or someone) forks the project,
I'd like to contribute something there as a developer and a translator.
Especially in the field of CJK problem solving.
Even if I can't it, at least, I can be the introducer the fork project and its products to Japanese users.

In reply to by knoike

I do have forked it (and do accept PRs to it, see https://github.com/Jojo-Schmitz/MuseScore/tree/master-to-3.x). But in my fork none of the GitHub CI (Continuous Integration) things exist, those are the ones building the artifacts, and also used to generate releases.

Setting up all that stuff, and possibly a separate site for distributing and supporting it, is something I currenlty don't fancy doing. So currently I just lean on the existing infrastructure.

In reply to by Jojo-Schmitz

I see. I understood that the your forked project is centered on code repositry.
I also understood that I need to build it myself to get an executable binary for the time being.

I have some question.

0.
For convenience, I think the your repository/branch may need a unique and easy-to-call name.
For example, like "FRJ (forked repository by Jojo)" or "FCJ (bugfix collection by Jojo)".
Do you have already any idea?

1.
Where is the issue tracker for that repository/branch ?

2.
May I distribute the executable file I built for testing purposes?
Especially, to Japanese users.

In reply to by knoike

Every developer of MuseScore works on his/her own fork (only those that have write access to the MuseScore would not need to, but generally do too).
And no, you don't have to build it yourself, as PRs against the MuseScore repo are build for you (or me in the case of that {PR #9000 for a 3.7](https://github.com/musescore/MuseScore/pull/9000)
You would hoewver need a GitHub account in oder to be able to see and download the CI artifacts

0.
My fork is https://github.com/Jojo-Schmitz/MuseScore. The branch with those changes for 3.7 is https://github.com/Jojo-Schmitz/MuseScore/tree/master-to-3.x, as I created a PPR from it (https://github.com/musescore/MuseScore/pull/9000), I can't rename it anymore

1.
My fork/repo doesn't have an issue tracker, because I haven_t created one. Not sure whether I want to...

2.
Yes! You can distribute your own ones as well as the artifacts from my (or anyone's) PR, there's nothing that could prevent you from doing this

In reply to by Jojo-Schmitz

Would a 3.7.Jojo be possible outside of musescore.org, e.g. a 64-bit Windows installation available on Github? I have seen the Github artifacts for 3.7 but I don't know how to install them. MS3 clearly has much life in it yet so it would be really nice to see bug fixes and security patches. Maybe it could continue for many years and become "Musescore Classic". Surely this would not need official approval.

In reply to by knoike

I was just looking for an easy upgrade path. I am not familiar with CIs, PRs and nightly builds but will try to figure out what to do with Jojo's mention of using bin/musescore4.exe to update my 3.6.2 installation. Perhaps there are detailed instructions on Github somewhere.

In reply to by knoike

From https://github.com/musescore/MuseScore/pull/9000 to "Show all checks", then to "Details" of the platform you're interested in, here the latest for Windows: https://github.com/musescore/MuseScore/actions/runs/3980051855/jobs/682…, there to "Summary" https://github.com/musescore/MuseScore/actions/runs/3980051855 and there you'd find the artifacts for 64-bit, https://github.com/musescore/MuseScore/suites/10512391754/artifacts/521… and 32-bit, https://github.com/musescore/MuseScore/suites/10512391754/artifacts/521…

That gives you a .zip file, inside is a .7Z file, inside that a MuseScore-3-7-0-xxxxxxxxx directory, extract that, inside that go to bin, there you'll find MuseScore3.exe.

Yes, it is a bit of a pain...

In reply to by yonah_ag

I could reproduct it.
No plugin required to reproduct.

[Plugins] -> [Plugin Creator]

In the Plugin Creator
[File] -> [New]
[Run]

Running…
Creating component failed
line 1: plugin cannot be loaded for module "QtQuick": Cannot load library C:\Users\ (account)\Downloads\MuseScore\MuseScore_x64_4104872289\MuseScore-3.7.0.4104872289-x86_64\MuseScore-3.7.0.4104872289-x86_64\qml\QtQuick.2\qtquick2plugin.dll: 指定されたモジュールが見つかりません。

Tested Env.
OS: Windows 11 Home 22H2
MuseScore version (64-bit): 3.7.0.4104872289,
revision: 095932d

It did not occur in Ver. 3.6.2. it ran successfully.

In reply to by Jojo-Schmitz

I've extracted the x86 artifacts and plugins now work. :-)
Now it's so easy to apply correct guitar "Let Ring" that I probably won't bother with my Excel plugout.
Thank you for the note length fix in PRE and plugin API, this really improves my workflow.

Note: The plugin uploaded above has a bug which stops note length adjustment - Ooops!
I'll upload the fixed version to GitHub with a few enhancements just in case anyone wants to use it.

In reply to by knoike

That C:\Users\ (account)\Downloads\MuseScore\MuseScore_x64_4104872289\MuseScore-3.7.0.4104872289-x86_64\MuseScore-3.7.0.4104872289-x86_64\qml\QtQuick.2\qtquick2plugin.dll looks quite long and has MuseScore-3.7.0.4104872289-x86_64 in it twice?
And that message doesn't match the one from the image you shared earlier

Try with reducing the length, by copying/moving that last MuseScore-3.7.0.4104872289-x86_64 toC:\Users\ (account)\Downloads` and run from there

In reply to by Jojo-Schmitz

It doesn't seem to be a problem of path length or location.

Running…
Creating component failed
line 1: plugin cannot be loaded for module "QtQuick": Cannot load library C:\MuseScore-3.7.0.4104872289-x86_64\qml\QtQuick.2\qtquick2plugin.dll: 指定されたモジュールが見つかりません。

Jojo, please try to manipulate following procedure for artifact MuseScore-3.7.0.4104872289-x86_64:

[Plugins] -> [Plugin Creator]

In the Plugin Creator
[File] -> [New]
[Run]

In reply to by Jojo-Schmitz

I tried the following.
It works fine.
I think the cause is in "qtquick2plugin.dll" included in artifact.

copy
qtquick2plugin.dll
from
C:\Program Files\MuseScore 3\qml\QtQuick.2
to
C:\MuseScore-3.7.0.4104872289-x86_64\qml\QtQuick.2

[Plugins] -> [Plugin Creator]

In the Plugin Creator
[File] -> [New]
[Run]

Running…
Plugin Details:
Menu Path: Plugins.pluginName
Version: 1.0
Description: Description goes here
Requires Score
Debug: hello world

In reply to by Jojo-Schmitz

Jojo, excuse me, please try following procedure.
If that situation doesn't give you any error, it is strange.
I'm guessing that the Jojo's environment is loading the DLL from another location.
(It's probably loading from the folder set in the environment variable PATH for SYSTEM/USER.)

1.
In MuseScore-3.7.0.4104872289-x86_64\qml\QtQuick.2,
rename "qtquick2plugin.dll" to "qtquick2plugin.dll.5_15_2".
(or delete "qtquick2plugin.dll")

2.
Run MuseScore-3.7.0.4104872289-x86_64\bin\MuseScore3.exe

3.
Manipurate MuseScore 3.7.0
[Plugins] -> [Plugin Creator]

In the Plugin Creator
[File] -> [New]
[Run]

Get an error as follows:

Running…
Creating component failed
line 1: module "QtQuick" plugin "qtquick2plugin" not found

NOTE:
"qtquick2plugin.dll" is not exist
-> "not found" error.

"qtquick2plugin.dll" is exist, but cannot load
-> "cannot be loaded" error.

In reply to by Jojo-Schmitz

Maybe I am wishing too much, but as Donna Summer said: Dreams come true for those who dream.
I am using MU4 because of Muse Sounds, specially Muse Strings. I like most of the other features of MU3 better than MU4. Would it be possible to borrow some code from MU4 into MU3 playing system to allow it to play Muse sounds too?

In reply to by Jojo-Schmitz

Maybe I am wishing too much, but as Donna Summer said: Dreams come true for those who dream.
I am using MU4 because of Muse Sounds, specially Muse Strings. I like most of the other features of MU3 better than MU4. Would it be possible to borrow some code from MU4 into MU3 playing system to allow it to play Muse sounds too?

In reply to by Jojo-Schmitz

This idea is OK. There's no need to start a new site now. It would get a long time to attract users. Keeping the information and talks at musescore.org and github is enough. And keeping Musescore classic side by side woth Musescore 4 is also enough. Users could jump from one version to another without as needed. In my case, I write many things in 3.6 and when I need a feature of 4.0 I "save as" my score with a new name and go on in 4.0. At the end it is just a matter of mixing the audio results in a DAW.

In reply to by fernandoamartin

No! Please don't do this.

I'm sure that MS4 will eventually have the missing MS3 features, and porting such a huge change is likely to impact MS3's stability. I think that MS3 (MuseScore Classic) should remain the pinnacle of MS's .SF2 based notation systems and move to a maintenance model. If users are really keen to see extra features then the plugin api would provide a route. Improved sound can be achieved by using exported MIDI with a DAW.

In reply to by fernandoamartin

no, Muse Sounds are a proprietary, binary-only feature that deliberately undermines the Open Source-ness of the rest of Mu͒seScore and indicates removal of any attempt of the current team to improve the base code’s audio capabilities (in fact, they’ve documented they removed some features, with mu͒3 as base).

In reply to by mirabilos

There's a muse-hub.COM site that suggests that they will sell plugins in the future. I understand it's their right since they are creating the plugins in their own costs. However many features existing in 3.x were removed and nobody knows how many of them will return. Some like the very limited sfz support were announced to be removed. Others like bringing back choice of presets in soundfonts are promised to return. After all, if we need some feature from 3.x the only way to secure it is maintaining it.

In reply to by Jojo-Schmitz

Jojo-Schmitz,

Respect and Thanks. I've just D/L and run 3.7-64bit linux version (aka 3.6.3 "Musescore Classic") appimage from Github, and it works well, so far, including Jack audio and midi. (Which is why I tried your version. I can't see Jack being added to 4.0 linux in the near future, if ever.)

Alex.

In reply to by Jojo-Schmitz

After so many months improving MU 3.x why not change the software title? Instead of keep calling it 3.7 unstable indefinitely, it could simply be Musescore 3 Rolling Release or simply Musescore 3 Rolling. That would really reflect the rolling character of the work that is being done cuurently.

In reply to by fernandoamartin

Actually, 4.1 should be renamed, since is a discontinuous fork, with a somewhat different feature set, maybe to something like "MuseCompose" or "MusescorePlus", reserving "Musescore" for the 3.xx line. That way there would be no ceiling on version number. It's discouraging to have to explain about the versions when promoting Musescore to friends and colleagues (although the 4.xx compatibility in 3.7 makes it less of an issue).

In reply to by jake11561

??? Why should the current version of MsueScor ebe renamed? It's not a discountiguous fork at all - MuseScore 4 is the successor to MusezScore 4 was to MuseScore 3 or 3 was to 2 or 2 was to 1. It's as plain and simple as that,. and confusing your friends and colleagues by pretending there is more to it than that is doing them a huge disservice. There is no reason anyone not already married to MuseScore 3 because of existing scores should be considering it at all. It's great there the 3.7 builds exist for the sake of those existing users unable or unwilling to update, but there is no reason any new user should even consider using MsueScore 3.

In reply to by Marc Sabatella

I can't in good conscience recommend 4.1 to people, knowing they are likely to encounter the playback bug(s) in 4.1, which I assume will get fixed eventually, but that for now, they should install 3.63, but, oh, there is a 3.7 which is the greatest version at the moment, but you have to get it from GitHub, etc. And they can't use my Plugins in 4.x. So it's hard for me to view 4.xx as a linear upgrade to, or superset of 3.63.

In reply to by jake11561

All releases of all software have bugs. Which specific bugs in MuseScore 4 do you think your friends and colleagues are especially more likely to run into that the average user will, and what makes you think they won't be more inconvenienced by the bugs in MuseScore 3 that would encounter instead?

In reply to by Marc Sabatella

Just going by my own experience. If said friends and colleagues ask me, all I can do is give them the benefit of my experience. They're welcome to consult other sources of info. I tout the orchestral sounds of 4.1, in case that's important to them. But honestly, for me 4.x has been a distraction and a drain on productivity so far. I'm looking forward to future versions, but for now 3.63 ( - 3.7) meets my needs and enables me to churn out the scores and arrangements my bands need.

In reply to by jake11561

If you experienced a bug that you think people should know about, then hopefully you also reported it. So please post the link so we can understand and assist better.

Merely telling your friends about bugs you found but not actually reporting them - now that would be something that should not be done in good conscience.

Anyhow, again, it's extraordinarily irresponsible to be advising people to use older versions of software with thousands of known and reported bugs and limitations that have already been fixed when there are new versions available. Especially when much better their scores could look, how much better they could sound, how many new features they could be taking advantage of, and how much nicer the UI could be, if they only started using MuseScore 4 right away. Plus you are dooming them to a second learning curve when they do update that could easily have been avoided. This is also something one shouldn't be doing to friends, in good conscience.

Again, it's fine for those people already using MuseScore 3 to delay updating for whatever reason because of concerns over specific things they rely on for specific existing scores or whatever that they know aren't working right yet. But advising new users to go out of their to install older versions instead, just because you aren't ready to update, is a terrible idea that does them a grave disservice.

In reply to by Marc Sabatella

The most serious (for me) of the playback bugs is reported: github.com/musescore/MuseScore/issues/16875. The lack of plug-in support in 4.1 is a known issue. Either is a show-stopper for me. And since it's likely that anyone to whom I recommend Musescore is likely to come to me for help, they need to know that I may not be able to help them with 4, and that they won't be able to use my plugins in 4. And yes, I advise them of all the great resources available to learn and learn about the latest version.
It doesn't help that scores could look & sound better and that the UI may be nicer, if basic functionality is missing.

In reply to by jake11561

Ah, so if you know that all of your friends rely on loop playback on repeats and your plugins that for whatever reason can't be ported to MU4 (plugins most certainly are supported in general), then indeed, this could be a reason for that small subset of people to resort to using MuseScore 3. But that doesn't validate your initial outlandish claim about 4.1 not being the true successor to MuseScore 3. Of course it is, even if a tiny percentage of users are for whatever reason not able to take advantage of it yet..

As for "basic functionality" being "missing", that's a matter of perspective. While A handful of things in MU3 are indeed not present in MU4, the reverse is an order of magnitude more true. Tons of "basic functionality" present in MU4 is missing in MU3. So by using MU3. you're missing basic functionality like scroll bars, Muse Sounds, braille, adding and removing instruments from parts, hiding staves within instruments, grouping measure onto a system, copying lines, chromatic ornaments, tempo change markings, harp diagrams, VST plugins, aux sends, and dozens of other things that would be as potentially important to a new user as "loop playback on passages that include repeats". Also you have to put up with vastly inferior score layout. It's an awful lot people have to give up just to have loop playback work with repeats.

In reply to by Marc Sabatella

Re: "plugins most certainly are supported in general", the most recent word from github.com/musescore/MuseScore/issues/8859 is that "only the plugins packaged with MuseScore 4.0 will work". And that's been my experience so far with 4.1.1

As I said, my use of MuseScore depends on those features, and I can offer advice only on things that I know directly. But I promise that I will tell anyone who asks, that v4 is THE official "Musescore", and to carefully check out fully all the features of any software that they may contemplate using. But if they ask me how I managed to change clef of a score in minutes during rehearsal, I won't lie about using v3.63.

How you decide to name your software is of course not up to me. I was just making a suggestion that I thought could provide clarity to new users, until v4 is fully developed (ie, can do everything that 3.63 does). Personally I have no problem navigating the current landscape. I didn't think that having two apps under the MU umbrella for some period of time would be such a sensitive topic.

It may be that v4 is so much better than 3 overall; you obviously have a broader view than I, so I will defer to you on that point. All I can tell you is that when I tried using v4, I immediately hit these snags, and was not inclined to pursue it further. My scores are not that complicated, so I'm not sure about the "vastly inferior score layout", all I know is that 3.63 does what I need it to.

In reply to by jake11561

The comment you cite about plugin was about the state of the program prior to the beta, several months before the actual release of the program. At that time, those were the plugins known to work. But very quickly after that, the community stepped up and got tons more plugins running, so by today there are about as many for MU4 as there ever were for Mu3. It's always been the case that the only ones actually supported by the core team are the ones shipped with the program - and there are more such plugins supported for MU4 than for any previous release.

You don't say which bugs you ran into other than that one involving loop playback (a feature not used by the majority of people), or if you opened issues for them. But note there have been four releases since the original 4.0, with hundreds of bugs reported by users fixed since then. So even if you ran into issues months ago, you may well find them already solved.

The vastly inferior layout of MU3 affects simple scores as much as complex ones. Just really basic elementary things like, on any given system, all eighth notes should have the same amount of space after them, same for all quarters, etc. MU3 consistently gets this wrong over and over. That plus terrible stem lengths and beam angles and more. Unless your music is nothing but whole notes, I guarantee it looks infinitely better in MU4.

Anyhow, again, everyone of course has the right to decide for themselves if some particular bug or missing feature means they need to forego all the improvements of subsequent releases (and thus putting up with hundred more bugs that have already been fixed, and dozens of missing feature not present in MU3). Everyone can make that decision for themselves based don't he specific features they care about. For people that care more about the behavior of loop playback in the presence of repeats than any of the other hugely important and significantly improvements I listed, then they are of course welcome to continue using MU3 for as long as they want.

I only stepped in to combat the misinformation about the nature of MuseScore 4 as anything but the official successor to MsueScore 3. It absolutely positively is, and for 99% of users - and virtually all new users not already dependent on older versions - MU4 is the no-brainer superior choice. It does a disservice to the community to suggest otherwise, and it's important that such misinformation be corrected.

In reply to by Jojo-Schmitz

A whole bunch of things in that list aren't actually missing features, just things you now do in a different way than before. But more importantly, that list was already out of date by the time 4.0 was released, and it's more out of date today. The list of MU3-isms "missing" from MU4 gets shorter by the day, while the list of MU4-isms "missing" from MU3 gets longer and longer and longer.

In reply to by Jojo-Schmitz

So like I said, if any given person feels the need to continue using MU3, that's fine. My issue was only with the statement about 4.1 not actually being the official successor to MuseScore 3 and advising all others not to use it without even first attempting to ascertain whether they would be more adversely affected by something missing from MU4 than they would by everything missing from MU3. That's just silly talk at best, irresponsible at worst.

In reply to by Marc Sabatella

Fine, except I never said anything about the "official" successor - all I did was offer a suggestion about nomenclature that could have clarified a path for new users and maintain continuity until such time that a fully integrated product was ready. But I promise henceforth to advise anyone who asks: Try the latest version, only come back to me if they can't accomplish what they want (and in my community of trad musicians, that's likely to be ABC import).

In reply to by jake11561

You called it a "discontiguous fork". That is saying it's not the official successor. And giving it a new name would do the exact opposite of clarifying things for new users. Right now, it's 100% clear: MuseScore 4.1 is the only current supported version of MuseScore, period. Giving it a new name would give the entirely false and harmful impression that it is not in fact the version they should be using, when it absolutely positively unequivocally is for virtually all new users.

People with specific needs who understand the tradeoffs and the enormity of what they give up are welcome to seek out snd install older versions or unsupported third-party forms like the one being discussed on this thread. But new users should under no circumstances be misled into thinking they should even consider this for a moment. Every single new user needs to understand plainly: 4.1 is the official supported version of MuseScore, period. Any suggestion otherwise needs to be countered to avoid causing harm to the MuseScore community.

Also, again, the ABC import plugin does work, in the way I described. Better would be a standalone web app.
But another possibility to consider - use 3.6.2 if need be just to run the import. Then save the file and immediately load it into 4.1 for any real work.

In reply to by Jojo-Schmitz

Just went back over the (many) past discussions on ABC import in the forum. ABC ImpEx crashes 4.1.1; I also tried JT's, but no success yet. I haven't given up, but I have to be conscientious about my time. And for me it's academic anyway until Loop Playback works.

Also, I think I need Plugin Creator to continue to develop/debug plugins.

In reply to by jake11561

Plugins that appear to crash MU4 are normally just executing "Qt.quit()" when they should be executing "quit()". So it's not actually crashing; MuseScore is just doing what the plugin asks.

But no, you don't need Plugin Creator to build plugins. It's sometimes useful for sure, so you can still develop them in MU3 if you wish to take advantage of it, then deploy them in both MU3 and MU4.

While most plugins will work with just that change (also the way they are added to the plugin menu is different), the functionality of opening or creating new scores does not work directly. So some plugins would need to be re-thought in a way that doesn't rely on this. That's what the current ABC import plugin does - saves the MusicXML and then prompts you to open it yourself. Not ideal, but it works.

Since the ABC plugins are really just calling an external web service to convert to MusicXML and then loading the resulting MUsicXML, the better solution here I think is just to create an actual standalone web app (eg, as a Chrome extension, or use github.io) with its own UI that uses this same service. It would be less limited in functionality and would be useful to people other than MuseScore users. I'm surprised there isn't something like that out there already. There did used to be a pretty simplistic online form to do this, but I don't see it anymore, and it could be done much more nicely.

In reply to by Marc Sabatella

But it's good for users to know that bugs have more chance of being addressed if reported at github. I found a bug of muse hub being unable to play microtonal intervals correctly, commented at MU forums at the bug report area, received some answers but nothing was done. Recently I posted it in github, along with others who had done so, and now it's fixed t nightly.

In reply to by jake11561

Revealing though it may be, this whole diatribe about MS4 vs MS3.6.2 is off topic for this thread so I have flagged all the contributions as spam. Please open a separate thread if you wish to continue your feud so that interested parties can also contribute without further clogging up this important thread.

In reply to by Marc Sabatella

It's completely off topic. It's cluttering up the most important thread relating to Musescore 3 Evolution and since it contributes nothing to the 3.7 fork then, in my opinion, it definitely is spam and is disrespecting this particular thread.

Please respect the MS3.7 community members by leaving this thread for its proper purpose. It would've been more than sufficient for a short, initial reposte: there was no need for the unhelpful mini essays which followed.

Please refrain from replying to this comment with yet another mini, (or even maxi), essay.

In reply to by yonah_ag

Again, the title of this thread is "End of life plan for 3.x", not "information about third-party fork of MuseScore 3". The end of life plan for 3.x is exactly this: MuseScore 4 replaces it. So pointing that out is not even the slightest bit off-topic.

If you want to discourage others from spreading misinformation about MuseScore 4 in that this thread, be my guest.

But if people spread misinformation in this or any other thread, it will be countered. And you're right - my initial response should have been enough. It's not my fault people continued to argue against the obvious facts.

So again, do NOT report posts as spam that are merely correcting misinformation. The misinformation itself wasn't even spam; it was merely misguided.

One more, to be perfectly clear: DO NOT report posts as spam when they are not.

And don't get on my case for "cluttering" this thread. You are the one who started this pointless digression by inappropriately flagging posts as spam. I have no intention of responding further here unless you or someone else spreads further information or takes further inappropriate action. The ball is in your court; I suggest you act by not acting.

In reply to by Marc Sabatella

More off topic spam. You are the one who felt the need to swamp this thread with an MS4 vs MS3 debate. You should've made a short initial reply, started a new thread and made a link to it. The thread would then have its own title on the lines of "MS4 is the official, supported version". Pleased to hear that you won't be posting another spam mini essay.

In reply to by yonah_ag

Again, I am not the one who is "swamping" anything. I am simply responded to misinformation and to malicious attacks. if I stop seeing misinformation and malicious attacks, I will cease responding, simple as that. Again, the ball is back in your court. Invite a response, though, and I will continue to oblige by providing one.

In reply to by Marc Sabatella

Honestly, I'm surprised with all this ado about MS3 vs MS4, as the discussion, even if it contained malicious misinformation (which i don't believe), is quite unlikely to cause any harm to MS4. This thread is actually a tiny and insignificant niche against (o parallel to) a formidable advertisement action carried out by a powerful company (Muse). A simple google search for "MuseScore 4" yields 379000 hits while for "MuseScore 3.7" only 803. But "End of Life plan for 3.x" yields a mere 4 hits! So the real impact of this discussion on MS 4 is vanishingly small!
Even if a particular user started using MS 3.7 instead of MS 4, It is inevitable that sooner or later they will try 4.x (it's in the ADN of all software users, isn't it?). If 4.x happens to provide them with a better user experience, they will migrate. Otherwise they will go on using 3.7. If they are good with that, what's the problem?
It's no use engaging in harsh and passionate discussions like some that I have read here for such an abstract problem.

In reply to by fmiyara

The damage of the misinformation and malicious attacks is not to MU4 itself, but to the community. It creates confusion in the eyes of new users who are erroneously led to believe there is some actual sound reason for them not to choose MuseScore - or worse, to choose an older unsupported version just because some small percentage of existing users need to use that older version for whatever reason. Which leads to people missing out on the better experience that they could be having - better layout, better playback, better usability - and creating an unnecessary extra learning curve when they do finally update. It also creates an environment where scores cannot be shared as freely due to compatibility concerns. because the FUD is causing both new users not to adopt MU4 and existing users who have no reason whatsoever not to update to continue using MU3 unnecessarily. And so on.

So once again, nothing wrong if some particular MU3 user who has carefully compared decides that some specific feature of MU3 is more important for their specific use case than all the new features and improvements in MU4 combined, But it's a problem when statements are made that are designed to do nothing but discourage innocent new users from getting the latest and greatest version of MuseScore.

And it's a huge problem when one commits personal abuse by reporting on-topic informational posts from a long-time contributor to this forum and to the software itself as spam. That was inexcusable, and what I expect to see here is people soundly condemning the perpetrator. The fact I don't is quite disappointing - even more damaging to the community than the misinformation.

In reply to by Marc Sabatella

Re: "And it's a huge problem when one commits personal abuse by reporting on-topic informational posts". Your posts are not on-topic so my flagging them as spam is not abuse and it's certainly not personal: please stop being so dramatic. I recognise and respect the fact that you have been a long-time and valued contributor to this forum.

"Innocent new users" are most unlikely to even find this forum, let alone this thread, they will just get on with using the software. So please leave this thread alone unless you have a contribution to make regarding MS3.7 - also known as Musescore 3 Evolution.

In reply to by yonah_ag

All users can absolutely see new posts here. If this were a private discussion, that would be different.

Spam is not defined as "anything you don't want to hear", and reporting posts as spam is an incredibly harmful act. You are engaging in abusive behavior, and every post of yours that doesn't apologize for that abuse is simply compounding it. I am sorry to see this community come to this. But anyhow, I have unsubscribed to this thread. If you'd like to apologize, you'll have to do so offline.

Crash report, sort of—
On Windows 10, the 64-bit flavor of build ca30ecd crashed every time I tried to open the demo files 'Brassed_Up' and 'Dynamic_Strings,' without any message or dump file etc--just crashed and disappeared. (It did successfully open 'Dawn.') Before trying to open other demo files, I switched to the x86 flavor, same build, which had no trouble opening all three files. I then switched back to the 64-bit and now the same demos opened just fine.

Just guessing here, but I wonder if the problem had to do with the dialog that asks whether you want to use Leland and Edwin. When the 32-bit presented me with that dialog I checked the box for applying those fonts automatically to older files. I noticed that when the 64-bit opened 'Brassed_Up' and 'Dynamic_Strings' they were showing the asterisk indicating unsaved changes. I assume both flavors are using the same settings—and that because I had set the default in the 32-bit, the 64-bit bypassed the dialog, and applied the font updates without having to ask.

(Which leads to a question: Where are settings like that stored? Is it possible to have different settings for different development builds/versions, and if so, how?)

In reply to by Stephen Cummings

I can't reproduce, loading these scores in my self-built 64-bit 3.7.0, which should be e9183eb, from today.

That setting is in Edit > Preferences > Import

I can't recommend that automatic conversion though: it also looses the Jazz style font (like in Brassed_Up)!
It really should only replace Emmentaler with Leland (and Free Serif with Edwin), but replaces all fonts.

In reply to by knoike

It might be related to the plugin issue mentioned further up, as that dialog is qml code and sure enough uses Qt Quick too.
Would explain why the 32-bit version is not affected and why the 64-but version is not affected anymore once that dialog got disabled via Preferences. And why I can't reproduce this too

In reply to by Jojo-Schmitz

Hi Jojo

MS3.7 gives this warning (Linux)
[....] Sample(Trombone F5) start(0) startloop(12406) endloop(12659) end(12666) smaller than SoundFont 2.04 spec chapter 7.10 recommendation
[...]
This with the appimage as well as my own build.
Warning from sfont3.cpp
I ignore if it there are any consequences because of this.

In reply to by graffesmusic

I think it is cause of SoundFont you use.
In the specification of SoundFont 2.04, there is the following sentence.

> Thus dwStart must be less than dwStartloop-7, dwStartloop must be less than dwEndloop-31, and dwEndloop must be less than dwEnd-7.

In SoundFont you using:
endloop(12659)
end(12666)
->
12666-7 = 12659. the 'endloop' value is 'equal' not 'less than'.
The 'end' value is only slightly smaller, so MuseScore may have showed a warning instead of an error.

In reply to by knoike

Unfortunately, almost all existing soundfonts violate that part of the spec; I talked to s.chriscollins about it, and he said it’s not been a problem in practice, and he’ll probably take care of these for new instruments but can’t bother for old ones.

Yes, it’s only a warning; FluidSynth upstream plays these fine, and the bastardised derivative thereof in use in mu͒2 and mu͒3 “probably” plays them fine.

It’s not a regression (change relative to a previous version) anyway.

In reply to by graffesmusic

ad 1: yes, they are the same soundfont basically anyway (one’s a subset of the other), but others are also affected

ad 2: that’s because the logging is either not present there or hidden as a nōn-development build; the problem has always existed, but the code that outputs the warning is recent (I wrote it as part of fixing some soundfont-related crashes)

In reply to by Jojo-Schmitz

Just bumped into this upload failure myself and then noticed these comments so I simply did the upload again via 3.6.2.

3.7.0 is a game changer for me because of the limit on PRE and plugin API maximum length being increased from 2000 to 60000. (64000 would be actually be even better because a 1/64th note could be made to ring for 4 beats of 4/4 time. It won't affect me in practise because I don't have scores with notes shorter than 1/32nd but it would just feel 'tidy'. I'm nit-picking – I really appreciate 3.7.0).

Maybe there needs to be a forum page for MuseScore Classic.

(or even MüScore Classic if the MuseScore name is not allowed).

In reply to by yonah_ag

> Maybe there needs to be a forum page for MuseScore Classic.

I think so, too.
For the time being, It may be a good way to create a new topic every problems/issue with "MuseScore 3.7.0:" at the beginning of the title as you did in the "Development and Technology Preview" Forum.

In reply to by Jojo-Schmitz

But it is fun to imagine possible names! Something from the repeats or endings palette may be appropriate: VoltaScore, CodaScore, FineScore, ...

Anyway, it looks like interest in this version is growing and the licensing model means that we're not going to lose the source code no matter what.

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