Continuing Problems with Musescore 1.0.
Dear User Group,
Since my last post, I've been having continuing problems with Musescore 1.0. and I can't go back to the previous version as the files are not backward compatible. The previous version was very reliable.
I have some work I did yesterday and whilst I can open the file, whatever I do just causes Musescore to crash.
Copy of file attached.
I have had other problems such as the cursor does not move at the correct rate on playback.
When I last reported a bug I was disappointed to get denial messages like "no one else has had this problem" and stupid comments about us of capitals in e-mails.
I hope you will take this post more seriously. I am a great believer in alternatives to Sibelius and have in the past spent a lot of time with an alternative product but I can't afford to do that all again unless I get more of a mature response from this group.
In any case, I am too busy to get into general discussions.
Tom Sutcliffe.
P.S. The Attachment is a first draft of teaching material. If you are an expert on the subject please don't find errors as it is incomplete and unchecked for the reasons indicated above.
I will now have to re-input all the work into Sibelius. Can't even create an xml file.
Regards.
| Attachment | Size |
|---|---|
| 4 Part Cadences .mscz | 3.94 KB |
Comments
I opened your file with mscore 1.00 under Ubuntu 10.04, and was able to add some notes, and to save
result is attached
In reply to no problem by robert leleu
Hi Robert,
thanks for sending back the file - I can open it but it still crashes whatever I do.
I can't see any notes added but I can see that the stems have disappeared from some of the notes. Did you notice that?
It just gets worse!
As previously indicated ( I should have said in the last e-mail) I am using Musescore under windows XP.
It is not much consolation to me to know that it works under Ubuntu if it does not work under such a common operating system as windows XP.
Has this version been fully tested under windows XP?
Tom Sutcliffe.
I could reproduce a crash on Windows Vista
- Open the score
- Click on the first E and go down half a step
- Runtime error crash
The same happens for all notes.
Tom, how did you manage to create this score in the first place if anything you do causes a crash?
In reply to Crash reproduced, MusicXML attached by Thomas
I don't get a crash on the score from Sydenham, but why do I get a pile of note changes on other bars? Many extra accidentals and courtesy accidentals appear when I select the first E and move it down.
The score uploaded by Leleu causes a crash when I select the first E and move it down.
Correction, try saving Sydenham's score as an XML and Musescore will crash. Where are all the rests that should be visible? Why do all the courtesy accidentals appear? The exported XML could be reloaded into MS but it justs worse from there. Sydenham's score is definitely corrupt.
In reply to Crash by schepers
in such cases I have been, sometimes, able to recover the work done by creating a new score and copying from the corrupted to the new one by small bits, saving at each step, in order not to loose the new score when some crash happens.
In reply to in such cases I have been, by robert leleu
This issue has been bothering me for some time as it still seems to happen too often. I've gone over and compared some of the internal XML (MSCZ) and the exported XML files (as they are somewhat different formats) in the hopes of being able to see where corruption might be, but the XML files are quite large and complex. Does there exist a (free?) tool to investigate and/or check XML files in the hopes of spotting corruption? It might help in determining when or where the corruption is happening.
In reply to Crash reproduced, MusicXML attached by Thomas
Hi Thomas,
As indicated in my previous e-mail,
it was working when I created the draft (all in one session) but kept crashing when I opened it again the next day.
I had had a lot of problems last time ( when I first down loaded this version) as indicated in my last post. It tries to look for an update on the net and then crashes - I got round this eventually by changing the settings to stop this happening. See my previous post for further details.
My the way I am using it under windows XP
These problems must be happening to a lot of XP users but most people will not bother to write and tell you about it. They will just not use it.
Tom
In reply to Crash reproduced, MusicXML attached by Thomas
Hi Thomas,
Sorry, I forgot to say thanks for the Xml file. I managed to open it in Sibelius, So I suppose I'll carry on with Sibelius.
How did you manage to do that? When I tried to save it as an xml Musescore just crashed.
The problem I have is, I've set up Musescore to load up the last file when starting so I can't get out of this situation with Musescore. Whatever I now do it just crashes. I suppose there is some ini file to edit.
Anyway, I think I've lost faith in Musescore 1.0 on Windows.
Regards,
Tom
The problem seems to be in measure 12, with the 2 8th notes and their beam. Find a cleaned version attached.
Thanks to have taken some minutes from your precious time to report the issue.
In reply to The problem seems to be in by [DELETED] 5
Lasconic, there's much more wrong with this score than just that bar. I load up your version, and just trying to move the first E down (in the first bar) using the keyboard down-arrow crashes MS. As Tom pointed out, there's notes missing stems, there's missing rests, etc. It's very messed up.
In reply to The problem seems to be in by [DELETED] 5
Hi lasconic
thanks for the the "clean version" I managed to load it in Musescore by telling it not to restore the last session but this version is corrupt too. It also crashes when I try to edit it.
There are stems missing from several notes (bars 12, 14 16 etc) and some of the accidentals are missing - for instance there should be flats in brackets in bars 12 and 13 (tenor part) and a non optional flat in the bass in bar 24 and there's a sharp missing in bar 25. In fact all accidental are missing after bar 6.
The bar numbers are also incorrect. what is shown as bar 12 is really bar 11 etc.
The only thing I managed to get it to do without crashing is to play back the score. It plays the missing accidentals but after playing bar 9 it appears to play a silent invisible bar before continuing to what should be bar 10. There is obviously something wrong there.
Fortunately I printed a copy before closing down the file so I know what was there.
Regards,
Tom
In reply to Thanks for the "clean" version. by sydenham
Here is a better cleaned version. Measure 10 is indeed the cause of the problem and not the beam.
Hope it works for you.
I'm very curious of why you choose to create this score like this with all these time signature changes, hidden rest etc... And why did you choose voice 1 and 3 instead of voice 1 and 2 ?
If I was asked to score it, I would have done it with Measure Properties -> Actual measure duration and the result would look like this file
In reply to Here is a better cleaned by [DELETED] 5
lasconic,
Thanks for that. The first file does correct the problem - but presumably you had to edit the file to do that. That's beyond my knowledge and not what I would want to do. I used to be a programmer and developer but I don't do that any more.
"Measure Properties -> Actual measure duration" is interesting but it's just another way of doing the same thing except the way you've done it some of the bars are too short. Remember, this is not a score it's a set of examples.
I was considering moving over from Sibelius to MS but the number of technical problems with version 1.0 make it too unreliable. Also, Because I use it for teaching material and analytical graphs I'm finding that MS is much too limited in many ways.
For instance:
- Manual formatting appears to be very limited: For instance: I can't find any way to more a system down exactly where I want it to make space for interpolated teaching notes and I can't find any way of getting rid of the "Piano" etc
- There is no facility to select an area of a score and create a graphic for inclusion in a word document for teaching or for webiste work or for use in a book. A serious omission.
- The number of graphics is very limited. A serious omission is the lack of a parallelogram beam tool that can be used for free format beams used in musical analysis graphs.
The above were all in Encore and are now in Sibelius. But all of this is a big subject that I don't have time to go into at the moment. When I have more time I may write suggesting new facilities I think should be in the next version.
On the other point, I suppose I used 1, 3 because they are shown on the tool bar in that sequence perhaps you should change the sequence from 1,3,2,4 to 1,2,3,4!
Regards,
Tom Sutcliffe.
In reply to Thanks by sydenham
Hi Tom,
Thanks for this extensive report. It's very interesting to define where MuseScore needs to work on, in order to match the level of product quality of Sibelius. Clearly we still have some work to do and your report is very helpful to identify this work. Thank you for that.
It's important to understand that MuseScore will not be able to compete 'yet' with a product that is already more than 15 years in the market. However we applaud anyone who is trying to replace their commercial software for MuseScore.
In reply to Thanks by sydenham
Actually, it *is* possible to do most of the things you mention.
- you can force a system down using a "spacer" (see "beams & spacer" palette"
- you can edit / delete staff names by right clicking a staff and selecting "staff properties"
- any number of third party screen capture tools can select an area of the score for conversion to graphic and inclusion in another document, so that lack of this feature in 1.0 is not really a huge deal - but FWIW, it is present in the nightly builds for the *next* major release
- while MuseScore has no drawing facility, you can creating graphics in any third party program and place them directly in your score
In reply to Actually, it *is* possible to by Marc Sabatella
OK but I think the priority is to get the reliability up.
on you specific points.
I managed to get the one about staff properties working. - Thanks.
The staff spacer - I've no idea how that works. I can't get it to do anything
I don't like mixing in third party software. it causes all sorts of problems and just takes ages to do anything. I think software packages should do what they should do.
screen shots are too low resolution for printed documents.
Thanks
tom
In reply to Thanks but . . . . by sydenham
The thing is, there are many thousands of users of 1.0 (not sure how many?), and there are almost *no* reports of reliability problems. So whatever you have run into here seems to be a fluke.
As for staff spacer, see the handbook section: http://musescore.org/en/handbook/break-or-spacer. Just drag and drop a spacer to the system above where you want the space, then double-click and drag the handle down.
As for third-party software, I would agree that in general, there can be interoperability problems. But graphics are about the most standardized thing in the world. There really should be no problems at all creating graphics and importing them. I agree that software packages should do what they should do - but what MuseScore should do is create notation, not graphics. So I wouldn't be expecting to see too much in this area any time soon.
You're right abut screenshots, though. So the facility in 2.0 to export selected areas will probably be very useful.
I should say that I also do a lot of projects that involve incorporating musical examples into word processing documents. By far the most flexible / easiest to use solution I've found is to create the examples in the ABC music language and use abc2mps to turn them into standard notation. This gives me complete control over the sizes and margins with more precision that I could possibly get by copying sections out of larger documents in MuseScore, or Finale or what have you. I wrote a set of macros for OpenOffice and Word to automate this process, so I can just type ABC directly into my document, hit a button, and have it turned into notation. I've been using this for all my teaching materials since the beginning of the school year, because I needed a way to make these available to a blind student (and ABC is pure text, so any screen reader can read it).
But this scheme does require a certain investment of time to come up to speed with ABC. If you're interested, check out http://abcplus.sourceforge.net/. And if you get that far, my macros are available on the Yahoo site for the abcusers mailing list: http://tech.groups.yahoo.com/group/abcusers/.
In reply to The thing is, there are many by Marc Sabatella
"there are many thousands of users of 1.0 (not sure how many?), and there are almost *no* reports of reliability problems."
This is just the kind of immature comment that puts people off reporting bugs and joining groups like yours. so the result is that less people report problems and so there appear to be less problems.
How many of the numbers you quote are Windows XP users? I am sure most of them would have the same problems that I have had with version 1.0. Pity as version 0.9 was much better. You are just in denial.
Most people who would experience problems would just give up and not bother writing about it. It is clear you have to be pretty resilient to report problems.
Unfortunately I can't spend any more time on this topic as I need to get on to other urgent matters.
I am very grateful for the many useful comments people have made and am generally impressed with what you have achieved with Musescore which is why I wrote in the first place.
Regards,
Signing off for the time being
Tom Sutcliffe.
In reply to Reliablity Problems by sydenham
"This is just the kind of immature comment that puts people off reporting bugs and joining groups like yours. so the result is that less people report problems and so there appear to be less problems."
No Tom, it's the truth. If you search through the forums, and most of the people responding to you (such a Marc, Lasconic, Thomas) are seasoned veterans or developers/coders of MS, you will see that 1.0 has been out for a few months now. It's popularity has increased greatly since 1.0 came out and your reported instability is not among the posts. I've only been using MS for a few months, and while it definitely has some issues to overcome, it is a reliable product under most conditions.
However, I'm surprised that no one suggested Tom do a "factory reset" to MS (you'll have to find the relevant posting on how to do that). Since you were using v.963, the old INI has been known to cause weird problems to 1.0. Many (most) of the people posting issues started on 1.0 thus started with a clean INI file. I suspect the format of the INI subtly changed between the two versions causing the problems, and I don't think the 1.0 installed removes the old INI.
In reply to Reliablity Problems by sydenham
Sorry, I didn't mean to come off as "immature". I am trying to be helpful here, and I am calling it as I see it. I have nothing to do with the development of MuseScore, but I have been a heavy user of MuseScore since shortly before 1.0 came out, and have been monitoring the forums pretty regularly. And I am being completely forthright when I say that my read is that the reliability of the program *is* very good. I mean, I'm not saying it's perfect, but if I were to make a list of the 10 or 20 most important things I'd like to see fixed based on my own experience and that of people posting here, none of them would have anything to do with crashes - they really *don't* happen particularly often as far as I can tell. There's a lot I'd like to see improved, but when you say say that the #1 concern is reliability, you are obviously seeing something very different form the majority of us here.
Anyhow, I have no ideas how many users are on XP, but I can tell you that the only significant difference between 0.9.6.3 and 1.0 is bug fixes. All evidence from my own experience and that of the posts here 9and what else could I possibly go on?) suggests that indeed, it is *more*, not *less* reliable than ever. I'm not saying you aren't seeing what you are seeing, but I am suggesting the overwhelming evidence suggests what you are seeing is a fluke. And the good news is, that means it will probably turn out to be a simple configuration problem (a reset does sound like a good idea). Or perhaps, one particular bug or small set of bugs you are running into over and over that others don't see, presumably because of something you do often that others don't tend to do much at all.
In reply to The thing is, there are many by Marc Sabatella
For what it is worth, I've experienced several reliability problems in MS 1.0. I've reported the ones I can reproduce, but there are some other weirdnesses that I had to work to overcome.
However, for the things I do (which are generally pretty basic) 1.0 seems much better than 0.9.6.
And I am speaking of files created in MS and only worked on in MS.
I have fixed some of the problems by using copy/paste to move music to a new, blank file. Since it appears that problems occur in files at particular measures, could a way be developed for users to detect that a measure has an anomaly, and allow users to "clean" that measure (even if it deletes all the measure contents)?
I am running Windows 7, so in general problems are not limited to XP. I haven't tried running in Linux, but I do have that capability so maybe I will do so for a while and see how it goes.
In reply to Thanks by sydenham
Once you found that Measure 10 was the problem. I just created another score and copy paste everything except this measure. I didn't edit the file manually.
- Manual formatting : You can use a spacer (See Break or spacer ) or a frame to insert space between 2 systems.
- To remove Piano, right click on a staff -> staff properties and edit the long and short names. See Release notes for MuseScore 1.0 , Important information for people upgrading
- A snapshot tool will be included in next version. It's already on the work. You can use a external snapshot tool such as Clip2net to do captures for the time being.
- I'm not sure what a parallelogram tool for beam is. I don't know this in Finale. I guess you are refering to Schenker Graphs. It's possible to change beams by double click. See Beam .
Let me highlight that MuseScore is a free and open source tool (it's not the case for the other tools your mention). Your contribution is very welcome if anything is missing. Still MuseScore is quite powerful and it's needed to invest some time to master it (as any powerful tool).
In reply to Once you found that Measure by [DELETED] 5
I welcome your comments but it is difficult investing time learning it if it keeps crashing!!
Anyway, most of my comments do not arise out my lack of learning.
Tom
In reply to Once you found that Measure by [DELETED] 5
I tried to do a Schenker graph I found on the internet. Just to see if MuseScore is up to the job.
J.S. Bach: Organ Chorale Prelude Wenn wir in hoechsten Noten sein (source)
In reply to Schenker by [DELETED] 5
I tried something similar but ....
I assume you are using the line tool for this.
The problem is this looks ok in low screen resolution but the joins to the stems are not correctly lined up in higher resolution. What is needed is a tool that forms vertical edges in the form of a parallelogram as in Encore or Sibelius.
The quality is not good enough for publishing.
Tom
In reply to YEs but by sydenham
I didn't use the Line tool here. It's only beams.
I think I understand what you mean regarding the joins between the beams and the stems. It's on the second staff , the first Z beam right ?
I still don't get the parallelogram stuff. Could you post an example done with Sib or Encore if you have a moment. Thanks.
Here is the score in MuseScore format, PDF etc... : http://musescore.com/score/5410
I'm an XP user and I've had no problems at all with MS 1.0. Conversely I had several problems with 0.9.6.3. My friend/collaborator found the same. But I think that probably only shows that PCs vary a great deal as does the way people use software. I appreciate that doesn't help with anybody's specific problems but I think it indicates that XP isn't the problem
In reply to I'm an XP user and I've had by fatwarry
[removed by poster - already posted in another thread]
In reply to I'm a Win7 user by leonbloy
Interesting - this seems to be the detailed reply you posted before and then it disappeared?
As for being able to select the components of a note separately - sometimes, that's a useful thing. For instance, to lengthen a stem. But I'd agree that your basic click in the vicinity of the note should select the note, not any single component. Maybe require a special sequence (eg, double click) to select just the flag? Doesn't appear you can actually *do* anything with the flag or stem selected except then double click or hit Enter to go to Edit mode, although I could be wrong about that.
In reply to Interesting - this seems to by Marc Sabatella
I think that's the basic argument people are having. Right now, a simple click-drag will dislodge the note stem or flag (or rests) and I see no need for that. A double-click will adjust the stem height, but I see no reason to be able to drag the flags, or rests or stems around the score.
In reply to I'm an XP user and I've had by fatwarry
I also find 1.0 very stable.
In reply to XP user by xavierjazz
and I find 1.0 very UNSTABLE
I had written a detailed reply in this thread, with steps and screenshots of my crashes.
I reedit the post several times and then a message came up "you have exceeded the number of posts" or something similar... and my post disappeared! (???) perhaps it is hidden for moderators to check it or something??
If not, I guess I'll have to retype it. I showed how I managed to crash my 1.0 installation, with my first created score, after entering just ONE note.
In reply to I'm using WIn7... by leonbloy
OK, but it's one thing to say it's possible to get it to crash, and another to say it's "unstable". Sure, any sufficiently complex program will have bugs somewhere that will cause it to crash, and you seem to have been unfortunate enough to run into one or two of these cases right off the bat. I note, though. that your examples appear to be triggered by trying to do something that shouldn't have worked in the first place (see my reply in your other thread). Obviously, that doesn't mean it should crash, and as a fellow user I'm glad you've posted the steps to reproduce these crashes so they can get fixed. But realistically, most people would never ever run into these cases, because they wouldn't be trying to do these things that shouldn't have worked in the first place. That's why these particular bugs have gone unnoticed until now, despite thousands of of people using the program every day - no one had ever tried to do the things you tried to do, because they weren't supposed to work in the first place. And that's pretty common in any software project. Handling of error cases - what happens when someone tries to do something that the program was never designed to allow in the first place - are always the most trouble-prone area of any application. So it often takes an inexperienced user go in and start poking around doing things that aren't actually correct things to be doing, in order to discover the errors that aren't handled well.
Anyhow, given that this is your first score, it's understandable that you'd leap to the conclusion that the program actually crashes a lot, but it really is *not* the case. Your particular crashes were caused by you trying to do something that isn't supposed to work, and that's the natural result of your unfamiliarity with the program. If you stick it out and start using the program as intended, I think your opinion will change radically.
In reply to OK, but it's one thing to say by Marc Sabatella
Fair enough.