MuseScore 3 Release Candidate

• Dec 22, 2018 - 11:15

Today we are pleased to announce the release candidate for MuseScore 3. Since the first beta release less than a month ago, we have resolved over 200 issues. Now, with all the most critical bugs fixed, we have produced one final pre-release build for testing. If no additional critical issues are confirmed, we will release MuseScore 3 on December 24!

Download MuseScore 3 Release Candidate

Still to come

  • Make MuseScore 3 available on Windows (32-bit)
  • Make MuseScore 3 available on MacOS <10.12
  • Finalize work on one of the most requested features: crescendo/diminuendo for single notes!
  • Rework plugin framework
  • Create public roadmap for MuseScore 3 development
  • Make video tutorials for all new features
  • Update handbook for MuseScore 3
  • Fix bugs, more bugs, all the reported bugs :)

Further reading


I hope that "release candidate" doesn't mean "this will be the first release, and the next will be in 6 months." Until to-me pressing bugs in fermata timing with 2.x scores and snapshot text scaling are fixed, this is not ready for my exclusive or default use.

In reply to by [DELETED] 1831606

I'll continue beta testing the "release" like I have been since the first alpha was released. I won't sell out to it until it actually works as well as version 2. There are still too many bugs still in the latest nightlies that keep it from being usable for all of my work.

In reply to by Jojo-Schmitz

It's about what I expect from a beta and 2 days isn't enough time to make it usable as a final release. Release candidate should mean the only changes to be made are previously unidentified critical bugs. You can sort of notate most music doesn't sound like a stable release for Notation software.

In reply to by [DELETED] 1831606

In reality, this version is not the "stable" version of Musescore 3.0.
After reviewing some application development sites, I found that the definition of "Release Candidate" (often abbreviated as RC) was a version of the software that was virtually candidate for the "final" or "stable" version of the application. She is available for "last minute testing" to identify the latest program errors.

In reply to by Jojo-Schmitz

Hi Jojo,
In IT terminology, a release candidate is supposed to be ready for production unless unexpected bugs are still discovered. While it must be balanced against the risk of unexpected bugs, a RC can already be used in production if a new feature is worth it.
Here, it is clearly not a release candidate but an additional beta called release candidate for marketing reason.
But it doesn't matter, what is important is all the good work that the team (including yourself of course) are currently doing, the program is becoming better and better in every beta, I will be very happy to use the new auto placement feature, thanks to everybody for the huge amount of work currently being done!

In reply to by Classical Empire

Does the terminology really matter? :)

We are developing great free product. It can be not stable enough for someone (as well as any of MuseScore X versions). What really matters is the new approach which allows updating very often. So, step by step we will make MuseScore stable enough for more and more people (including you, for sure) ;)

So, stay around and thank you.

In reply to by Anatoly-os

In a way it does. I just re-downloaded it.
But I usually wait for at the bare minimum 'release-candidates'
This really isn't one. It is a beta still.
But if big companies starts mushing the word, it will be useless.

Look at the damage FallOut 76 caused. I will never pay for a pre-release again.

In the grand scheme of things, yeah it doesn't matter. But release-candidates are rapidly becoming the 'luxury' in luxury apartments. An absolutely useless term that now means absolutely nothing. Have you ever seen just an apartment now?

Anyway, awesome software best of luck.

I loaded my test file "Bach_demo2.mscz" from node "MS3 Beta update (14.12) - syllable overlap" with MS3-RC.

Unfortunately, the result is still bad. Well, # 151256: "Long lyric syllables at end of measure can overlap first syllable of next" is still open. Note that the title is misleading. As dmitrio95 has already shown on the 16th of December, the overlap of syllables also occurs in the middle of the bar.

In its current state, the release candidate is therefore not operational for choral music.

I found, the change of spatium is only posible in steps of 0.2 mm. Other settings now are only possible by changing to inch dimension. Coming back to mm you get values that fall out of the 0.2 grid.
Is this a feature? I always like to get the biggest notes that fit into a given number of pages.

In reply to by SlyDr

In fact, you can change spatium even by 0.001 mm but this seems to work in not very convenient way. Try to select the digit you would like to change by mouse and type any other digit. That will allow to get any number you like so no size grid was probably intended.

Concerning the behavior of that input widget, I suspect that the reason of it may be on Qt side. But I don't know it for sure.

Crescendo/diminuendo for single notes is actually a crucial feature needed to be added. Keep on working, and nevertheless, congratulations to you all!

In reply to by jagree

True.. which is also why there is quite active development of at least two people happening around this subject. As it currently stands they won't have finished their task before the set goal for 3.0; but it is looking very promising that this will be made possible within the first half of 2019.

I was simply triggered by the specific choice of wording. Don't get me wrong, nearly all of my contributions to MuseScore have been about playback rather than notation as well ;)

A huge thank you to the team that does all this work. You are true heroes in my book and I can't thank you enough for all your efforts. You're making one of the best pieces of software (and I don't just mean "notation software," I mean software period) out there. Thank you and a very happy holidays to you all.

In reply to by John Asendorf

Lets put progress into perspective. This release candidate is more stable than the version 2 release candidate. Most things i have seen that are causing crashes you can work around. Any file corruption can be sorted with the debug tools which is a fantastic position compred to version 2 There are some features that may be issues for certain gen’re of music but all in all it is well on track to be a more stable and feature rich scoring package.

Thanks for the efforts folks, more than happy to post any new crash modes that hsppen

at present they have been all reported.

Merry Christmas

What about Musescore-3.0.1 or 3.0.2, during the change the languages in KDWin this programs enter
in conflict with KDWin and become not responding. Therefore I decided to return to 2.3.2 version. Let it without automatical placing but it so easier starting in my computer

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