Musescore 4.0.2. crashes on windows 10 depending on audio settings !!
Didn't use musescore for a few weeks, when started today after a splash screen the app crashed.
On this forum I found some people using sonar audio driver gives the same problem, I don't have this audio driver but using the audio from my mainboard (High Definition Audio Device).
So I tried another Audio device, I could user my monitor(s) as audio output, when choose one of them Muscore 4 started and did an upgrade to version 4.0.2.
This version also crashes with the standard audio output.
So I found a workaround but an App shouldn't crash on audio settings.
Some work to do for the MuseScore programmers :)
When I can help to test something, let me know.
Comments
There is a known bug that can cause a crash with certain audio drivers and it should already be fixed for the next update. A useful thing you could do is test a nightly build (go to Download / Software on this site, scroll down to access the development builds).
In reply to There is a known bug that… by Marc Sabatella
Hi Marc,
I tried the nightly build (latest) from 18.05.2023, version 4.1.0.dev and it seems the problem for the audio output from my mobo is solved, when I start Musescore with the audio output where 4.0.2 crashed, with this version It starts as it should.
In reply to There is a known bug that… by Marc Sabatella
Can I read more about this bug fix somewhere? I have a problem with Musescore 4 crashing with one of my audio devices (M-Audio AP2496) unless I disable that device before starting Musescore. Very inconvenient! I'd like to get a sense of whether or not the nightly build is likely to help with that before I install it.
In reply to Can I read more about this… by rednoise99
Nightly builds are not "installed". They are like a portable app. You can download the nightly, try it, and simply delete its folder when done. It won't interfere with any installed programs.
See:
https://musescore.org/en/nightly-builds
In reply to Nightly builds are not … by Jm6stringer
OK. I'll check it out. But isn't there a changelog saying what's new or changed in each nightly?
In reply to OK. I'll check it out. But… by rednoise99
Sure, more or less - just check the project history on GitHub. For example, this is a listing of all merged PR's from present working backwards: https://github.com/musescore/MuseScore/pulls?q=is%3Apr+is%3Amerged
Figuring out which specific PR fixed any one specific crash would take some work, only worth it to you if you would have some special reason to need to know.
In reply to Sure, more or less - just… by Marc Sabatella
That would probably be more work than it's worth, faster to just try the nightly. Which I just did, and no, my issue is not fixed. My problem is that MS4 crashes silently on exit, that is, the interface closes but the process lingers. Starting and exiting the program multiple times leaves multiple processes running, none of them which can be killed. The only way to get rid of them is to reboot. I've reported this but I have little hope that this will be fixed, since nobody else sees such a thing. It appears to be related to my audio interface, which is quite old and not made for Windows 10, but has worked with everything (including MS3) I've thrown at it until now.
In reply to That would probably be more… by rednoise99
Hmm, sounds like you are describing something entirely unrelated to the subject of this thread then. You say you've reported it - where, exactly?
In reply to Hmm, sounds like you are… by Marc Sabatella
Crash+Audio Settings. Sounds related to me.
I described it here a few weeks ago. Not a peep from anybody. Plus I've mentioned it in passing a couple of times, like this time. I'm considering getting a new modern interface.
In reply to Crash+Audio Settings. Sounds… by rednoise99
One is a crash on startup, the other on exit - pretty different really.
Anyhow, I see a post from you here - https://musescore.org/en/node/347374 - and you got a response pretty quickly but you never followed up, so I suspect the assumption was, the problem went away or you lost interest.
But it looks like you posted enough info for me to say it would be worth making a formal report on GitHub. This forum is great for getting feedback from other users to confirm if something is a bug, or if it's already reported, or if it's already fixed. But in the end, if something does turn out to be a clear bug, you need to open an issue on GitHub. Definitely best to check here first, but crashes are always bugs, so if you don't hear within a day or two that the crash is already known or already fixed, or that more info is needed in order to reproduce the problem, just ahead and head to GitHub. That's when the developers will be able to start investigating.
Unfortunately, there probably isn't a ton that can be done without clear steps to reproduce the problem, but maybe the details you posted will be enough to give the team good clues as to the problem.
In reply to One is a crash on startup,… by Marc Sabatella
OK, you win, I lose. If you say the issues have no resemblance, you may carry on as if I never mentioned it here. Which I did only in passing. My comment was really about the presence of a changelog, not to air my trivial problems (having to reboot every time I use the program.)
I didn't see that response to my previous message about it. Now that I do, I must say that it wasn't very helpful or friendly. In essence, he says, "I don't have that problem, so you should worry about other things." Well, gosh, thanks so much.
In reply to OK, you win, I lose. If you… by rednoise99
Relax, it's not competition, we're all winners here :-) The only way you can lose is to not submit the issue on GitHub, because that would mean the developers won't have a chance to investigate and hopefully fix it.
Anyhow, regarding your previous post, the point in the response was, I assume, to let you know you it isn't a problem affecting everyone and therefore we'd need more information from you in order to understand what's different about your system. It could just be that you are the only user with that specific audio device, or it could be something else - really still too soon to tell. But again, next step is GitHub
As for why I brought up the relatedness of the issue, it's just to point out that indeed it shouldn't be surprising a fix for the other problem didn't fix yours, and since you hadn't yet explained your problem was different, I was asking for more clarification. Again, not a competition, just trying to get the information we need in order to provide the best possible assistance we can for your specific problem.