Transcribing a tricky score - Musescore 2.0.3 to the test

• May 28, 2016 - 07:11

I must start by saying that I’m not actually a Musescore user. The program that I mainly use to write music is Sibelius, which I have been using since 2001. However, I've always had a lot of interest in new music notation software and I have been paying attention to Musescore’s development since its first version, when it was a program only available on Linux platforms. I doubt that I would switch to Musescore (unless Sibelius is discontinued). Don’t get me wrong, I’m in no way affiliated with Avid and Musescore is a wonderful program, capable of doing almost everything I can do in Sibelius, but the many features that have been automated in Sibelius allow me to work faster and with much more ease than Musescore.

Having said that, let me say that what Musescore’s developers have accomplished is outstanding. The first music notation software that I used was not Sibelius but Encore (it’s still around, surprisingly) and you wouldn’t believe all the things I could not do back then. Something as simple as making more space between two staves was impossible and dynamic parts was a futuristic dream (that eventually came true, thank goodness!).

As I said above, I have a keen interest on new programs and I like to toy around with them, especially if they are open source, like Musescore and Lilypond (which I have also used). Therefore, two weeks ago I decided to put Musescore to the test and transcribe an old composition, which I wrote in 2004, and see what the program was capable of. The music is pretty straightforward in its style; what I mean is that it's easy to grasp. The themes and sections are very clear and the harmony is very simple. However, the counterpoint is pretty elaborate and a few parts – especially, the violins – require some virtuosity. The score is not very complex but the middle section is somewhat tricky because it has no metre (i.e. no barlines and no time signatures) and a bit of contemporary notation is needed. I have uploaded the file to the Musescore site. Here’s the link:

//musescore.com/user/361186/scores/2142116

I won’t go into a long and detailed explanation of my experience transcribing this score on Musescore. You can see the result in the link above. I prefer to simply summarise a few interesting things I discovered:

1. There is no information in the online documentation (or at least I was incapable of finding any) about what things can be changed in the parts without affecting the full score and vice versa. I had to figure it all out by myself, trying things like making something invisible in the part and then seeing if it also became invisible in the full score (It doesn’t, by the way) and stuff like that. Of course, this is not such a terrible omission but I still think the documentation should explain in detail how parts are handled.

2. Frames in the score create fixed line breaks in the parts. I wanted to use frames to separate systems and place some texts (like footnotes) but it was impossible. Frames do not only appear in the parts (which they shouldn't, in my opinion) but they also break the lines in the same places as in the full score and there’s no way to fix it. In the end, I had to let go the idea of using frames at all and used spacers instead. For the texts I used Staff Text and then placed each by hand.

3. I was a bit disappointed to find out that all the careful work I had painstakingly done in manually resolving all collisions between objects was lost in the parts. I had to do it all again. “Oh, the pain, the pain!” Fortunately, it’s a short piece and it involves only six instruments. I would have gone crazy with a piece for full orchestra. I read that the solution for this comes in Musescore 3.0, which apparently will use “intelligent” algorithms to automatically resolve all collisions. Kudos for the Musescore team!

4. One thing that I missed a lot was the ability to copy and paste lines/spanners. It speeds up my workflow considerably. There’s a way of doing it by clicking Cmd+Shift (on a Mac) and dragging the object. However you have to do this one line/spanner at a time. You can’t make a selection and copy/paste them somewhere else. I wonder if this is something that could be solved in a future update of Musescore 2 and not have to wait until version 3.

5. There’s a bug in Musescore regarding beams that cross a barline. They look great until a line break occurs and the second measure is moved to the next line. The continuation of the beam then jumps up on the page and can’t be dragged down to it’s proper place. Take a look at the piano part on page 8 of my score. Don’t be fooled. There are no beams crossing the bar line where line breaks occur. I have carefully placed small lines with their width increased to 0.50 at the end of one line and the beginning of the next one. One painful detail here is that line breaks are different in the part so I had to do a lot of tweaking, placing lines on the score and hiding them on the part and vice versa.

6. One thing that I discovered is that the text in text frames can have a different format in the full score than the same text in the parts. Line breaks can occur at different places, for example. However, I then discovered that you could in fact delete a word in the part and the change would not be reflected in the full score. This means that these objects are not linked; something that must be corrected.

7. It would be a good thing to include a “special page break” function that would allow a user to insert a blank page in the full score or in the parts. In the full score, this helps to start another movement in an odd page (if the previous one also ended in an odd page). In the parts, this helps to solve problems with page turning.

8. To my greatest admiration, Musescore was capable of doing some stuff better than Sibelius. For example, there’s a bug in Sibelius were the accidentals of cross-staff notes are repeated on every occurrence of the note. You have to hide them one at a time (it won’t work if you select them all first) or use a plugin to solve the issue (which will simply hide the accidentals). Musescore also places mid-measure clefs correctly, while Sibelius doesn’t. I have to adjust them by hand all the time. Finally, slurs are better placed in Musescore than in Sibelius (although, to my taste, the shape is not good; too flat on the edges and too high in the middle). A slur that ends in a stem is not placed above it, like in Sibelius, but at ⅔ the distance of the stem, which is accustomed in music engraving. And the list continues.

On the overall, I think Musescore performs pretty well. Once version 3, with its “Smart Layout” feature is released, its users won’t have a thing to envy from their peers that use Finale or Sibelius, except good playback. Now, you may argue that this is a feature that a music notation software shouldn’t care of but since Sibelius and Finale did it, the rules of the game have changed. And the ability to create realistic mock-ups directly from a score writer is much appreciated on these days. It saves a lot of time!

Finally, I must say that I managed to do all that I needed to do. The score looks exactly as I wanted; even if I had to suffer a little to resolve all collisions by hand and to find a way to address the problem I mentioned in point 5. The only thing that I couldn’t do was adding title pages, but I managed to create them on LibreOffice and joined them with a pdf-handling software (however, they are of course not included in the Musescore file that you can see on the link above). The dynamic-part feature works pretty well and the parts look great. My congratulations to the Musescore team! I wish I had a program like this when I started composing!

Antonio


Comments

Other comments come soon. Meanwhile, thanks for sharing.
For the eventual cover,
Star from a new score and add a vertical frame;
Click the right mouse button on the second and add Title etc .;
Click the margin of the first and drag to increase it to occupy the whole front page.
Drag your image and drop it in the first frame.
see: https://musescore.org/en/node/112561#comment-507046

In reply to by Shoichi

Yes, I tried that, although I hadn't seen that thread (maybe because it's in Italian LOL). However, I couldn't find a way to hide the page numbers on the title pages and restart the count on the page where the music starts. (This is how I format my scores all the time and it is also a recommendation from most music editors. In addition, all the printed scores that I have show no numbers on the title pages and the page where the music starts is always page 1. ) But I must recognise that all this numbering requirements are not really important after all. I'm simply a very picky person when it comes to my scores. Most people wouldn't bother, I'm sure.

In reply to by Shoichi

You were right! It worked! I will upload the new score right away. Apart from having to hide all the texts in the parts and reduce the height of the frames (I chose to reduce them to 0.01 so they will take almost no space) I encountered no problems. The only thing I missed was the possibility of justifying the text of the programme notes. I hope they introduce text justification in a future version of the program.

Thanks for the comments A few notes:

1) feel free to suggest specific documentation changes, or make them yourself - that's one of the great things about open source software!

2) I'm not clear on why you need the frames with text in the score but not the parts - that wouldn't be common, which is why it doesn't work this way. But if you do have a need for this, then indeed, spacers will work.

3) The problem with having manual adjustments also affect the parts is that it is very rarely the case that the *same* adjustment works for both score and parts, because they are formatted so differently. BTW, it isn't clear yet just how much automatic collision avoidance will be implemented for 3.0, input on what you think the priorities should be is always appreciated (best to start a separate thread for this)

4) Spanners copy when you copy the measures. So you could copy the measures then delete the notes (using the Selection Filter, for example).

5) This a known bug, unfortuantely it has proven tricky to fix.

6) Can you file an official bug report (Help / Report a Bug, from within MuseScore) and attach a sampkle score and steps to reproduce?

In reply to by Marc Sabatella

Hi Marc! Thanks for the comments!

1) Yes, I know I can do that. I'll try to find some spare time to do it.

2) Oh, no. I needed them in both but the text frames also broke the lines in the parts on the same measures as in the full score and, in many cases, the line before the frame would end up with just one measure (and reformatting the previous lines didn't produce a good result either). I was referring to plain frames. By the way, what is their purpose, exactly? I was using them to make more space between systems in the full score. That's why I didn't want them in the parts also.

3) Yes, I suppose you're right. I see. Great idea! I'll make a list of all the things that I would consider important and create a new thread for that.

4) True, but it doesn't fit my workflow. I usually first lay on the notes and then start placing slurs, hairpins and dynamics. But that's just me, anyway. Don't get me wrong. I don't consider this thing too important. I'm afraid I have been spoiled by Sibelius but, if you could see what I had to deal with when I used Encore, back in 1998! Compared to that program, Musescore is the ultime tool in music notation!

5) Well, beams that cross bar lines are not a common occurrence anyway, so I wouldn't consider this a priority.

6) I will!

Oh, I just realised I forgot to mention one thing. The PDF output from Musescore looks blurry in both my iMac (with retina display) and my old LCD monitor. I guess it's better to create a thread for this too.

In reply to by Antonio Gervasoni

2) Without seeing the score in question, it's hard to imagine why you *wouldn't* want the frame to appear before the same measure - the text presumably has a function within the score and getting the same position would seem important most of the time. And that's the purpose of a frame - to create a spot for text or other content at a specific point in the score. I guess you have some very unusual special case in mind ewhere it might be OK for different people to see the sme text at different times, but I am having trouble imainging it, which is why seeing the score would help.

4) Instead of thinkng in terms of *copying* hairpins etc from part to part, just add them to all parts at once. Eg, select a range across multiple staves press "<" or "S" or double click an icon in the Lines palette, and the line is instantly applied to all staves at once, no copy/paste required.

FWIW, there are those who consider Encore to be the ultimate in usability and are constantly begging us to making MuseScore more like that. Different strokes for different folks! I too got into the notation software world around 1988, but for me the first mainstream program I used was Notator, which did a lot of things very well for its day.

In reply to by Marc Sabatella

But you can. It's in my post. Or isn't it? At least, I can see it. Can you see it was well? (Please let me know if you can't). Or do you mean the file itself? But in that case you can download it, right?... Or not?

2) No, I'm not talking about text frames. Of course, they should appear on the same positions, for obvious reasons. I'm referring to plain frames. I'm beginning to think that perhaps I was just trying to use them for something they have not been programmed for. I wanted to add more space between systems in the pages that had two lines instead of three, so that the bottom margin is the same in all pages (an old custom), and I was using frames for that. Maybe that's not what they are intended for. The solution was simple: I used spacers instead of frames. Voila!

4) I think this is just a case where my workflow conflicts with the way Musescore handles spanners. Let me explain: wherever the same dynamics (expression marks plus hairpins) have to be repeated on several lines (assume a long string of crescendos, diminuendos and different expression marks), I usually first create the dynamics for a single line, select it (all items in it), copy it and then paste it all around. I find that this approach is faster than creating the dynamics individually, even if I can select several lines and apply the same expression text/hairpin to all of them, simply because the act of making all those selections takes time. Of course, it's just a few seconds! However, once you gain a certain speed at performing a given task, it's always difficult (from a psychological point of view) to slow down. :-)

Oh, I was talking about Encore back in 1998. I don't know what the program can do today. Maybe its capabilities have been expanded and no longer has all those limitations (which, by the way, I didn't notice at the time). That would be a good thing, indeed!

In reply to by Antonio Gervasoni

It wasn't clear you were talking about the same score you already posted; I didn't see a place where you were using text frames in an unusual way that would require them to appear at different locations in the score versus parts. So, where specifically are you talking about? The footnote at the bottom of page 7? I guess I can see how that would be problematic, and that we would probably need a better way of dealing with that. But the way you chose is probably the best for now.

In reply to by Marc Sabatella

Oh, you wouldn't find any place because I removed all the frames. :-)
The footnote on page 7 is just Staff Text (which I had to copy on each part and then hide one by one in the score). I tried with a frame but it didn't work. The piano score would always show a single measure on the line just after the frame and I couldn't find a way to fix it.
Yes, it's obvious to me now that frames aren't supposed to be used the way I was using them. Spacers are right choice.

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