Change Instrument text doesn't stick on repeat

• Dec 15, 2017 - 19:48

I'd like to play a 32-measure sequence with a particular instrument; then change the instrument using the Change Instrument text on a rest (or if I have to, a note) in the last measure of the sequence; then have the new instrument play on the repeat. But the instrument change doesn't seem to stick for the repeat.

Is there any way to do this, short of duplicating the entire 32-measure sequence?

Thanks!

Attachment Size
Change Instrument for Repeat.mscz 5.71 KB

Comments

In reply to by Shoichi

Thanks, Shoichi, but no: I want the second pass of the first measure of the piece to be played with a Koto rather than Sitar. In other words, I want the instrument change applied in the first volta to persist into the repeat.

You cannon make one instrument play on one repetition and another on the next. The measure is affected by the instrument assigned to it last on that staff. An instrument change after a measure will revert back to the first when the repeat jumps back before that measure.

In reply to by mike320

>> The measure is affected by the instrument assigned to it last on that staff. <<

Which is exactly want I want, Mike; except that I want it to follow the virtual flow of the piece rather than the physical layout (which, to me, only makes sense).

In reply to by manonash

I was only answering your question, not giving an opinion. The last instrument on the staff in your example was the one assigned to the staff when it was created, not the instrument change in the Volta. I'm sure there are others that would find what you want useful also but I have not found a reason to use it.

Having said that, in Version 3.0 I don't think it will be possible to do it exactly as you are requesting (though I may be wrong), but will be doable. Since you have two instruments with no time to change instruments, it would make sense to write for 2 instruments in the score rather than one with an instrument change. What IS expected in version 3.0 is that you will be able to tell each of the instruments on which repetition to play, if you use two separate instruments. In 3.0 it may be possible to merge the two instruments into a single staff, which would make it possible to do exactly what you are asking. The bad news is that 3.0 is still quite some time off.

In reply to by manonash

Indeed, it makes sense. Right now MuseScore doesn't have any concept of playing things differently the first versus second time on a repeat. It's something that could be added some day, but the focus of development is mainly on notation, with playback being more secondary in priority.

In reply to by Marc Sabatella

>> the focus of development is mainly on notation, with playback being more secondary in priority.

With all due respect, Marc, I find that a false dichotomy. My compositional process -- and I'm sure this is true for many if not most composers -- involves continually hearing and re-hearing what I'm working on (right in the software). The notation and the rendering are inextricably interwoven.

When I was a software developer, one thing I learned early and learned well was that redundant sections or snippets of code were anathema. They are accidents waiting to happen. You clone a snippet, paste it somewhere else, maybe even modify it a bit; then later, almost inevitably, you find a way to improve the original snippet. Do you remember to change the copies and near copies? Do you even remember how many there are, and where they are? Do you know whether they are exact copies, or slightly modified copies? Do you remember what those slight modifications were?

It's a nightmare, and the same dynamic applies in musical scores. Copying an entire 16- or 32-bar section of music just so one can change an instrument, or add an instrument, or whatnot, is just a bad idea. Now you have to remember to make every change twice -- or three times, or four times, depending upon how many of those infernal copies you had to make. And you can't press forward with creative work, even when you're in a magnificent creative frenzy, hoping you'll remember to make the redundant changes later. Inevitably you'll miss some, and then teasing out the differences between the copies becomes a big job. Or, they jump out and bite you in the middle of a performance!

I write many kinds of music; but as a performer, I'm a guitarist, and am, more often than not, singing while playing. There are situations when I can turn a page during a song, but there are many more where I can't, so that economy in the score is extremely important. Three pages is the practical limit of what I can unfurl on a music stand, and I can't have notes and lyrics so small that I can't make them out in the less-than-optimally-lit situations that always come up. Unnecessary bloat in the score is a serious problem!

Making a second copy of a score -- one that doesn't play back correctly -- simply to achieve notational economy is another version of that deadly clone-and-modify strategy. How do I know when a piece is finished? Most of the things I write are never finished. The amount of tweaking becomes less and less over time, and approaches zero, but never reliably gets there.

So I say that in scoring music, as in writing software, things should be written for re-use whenever they can. In software, you add parameters to a routine to enable it to handle needs that vary slightly from use to use. In music scores, you can do the same thing by adding textual notes -- e.g., "2nd time only". But having one score for proper playback and a different score for economical notation? Not good!

In reply to by manonash

Indeed, listening to the playback of a score can be an important part of the compositional process. I'm just pointing out that when it comes to priority, getting the playback basically close enough to give you an idea of how things will sound is more important than getting every last detail right. It's impossible for a tiny team of volunteers to do everything, so we focus on what makes the difference in a notation program - and that's notation, plus whatever the basic playback is needed to at least give you a ballpark idea of how the piece will sound.

If you also feel like personally going the extra step to make a second copy of the score with even more realism for playback, that's totally up to you. I don't recommend it really, for exactly the reasons you state, but if it's important enough to someone, they are welcome to do so.

If we had more volunteers with software development experience and an interest on improving details of playback like this, then that help would of course be most welcome!

In reply to by Marc Sabatella

Well, I hear you. It's just that, for me, I compose many, many more pieces than I have either the players or the forums to present live; so the Musescore rendering is, more often than not, my only means of sharing a piece. Accordingly, the quality of playback in my notational software has become a greater and greater issue for me. One of these days it may drive me (ugh!) to Finale or Sibelius, if I learn that they can really do a much better job of playback. Or maybe the solution is midi export to a DAW, and lots of twiddling there. But that feels like a bit of a technical rat's nest, so I'm reluctant to get into it, having burned up so much time and productivity in a variety of those along the way.

As for participating in software development: well, I just turned 69, and spent four decades of my life doing software (and only a tiny slice of that doing anything like composing music) ... so software development is just not something I do anymore, in the time I have left (which reveals itself clearly as finite at my age). I do try to contribute by framing issues that I come across as clearly as I can (in which enterprise my software background makes a healthy contribution).

I do appreciate greatly the wonderful work that you and all the software developers for Musescore do. Musescore has made a wonderful contribution to my life. So, thank you!

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