Add "second time only" and similar playback options for repeats (not Voltas)

[Jolter]

Your idea

Very common in big band music to have a solo that repeats for example 4 times, and where some parts will play backgrounds only on the last repeat of each soloist, or similar.

Common notations I've seen in the wild

"Play first time only" or "1st time only" over the first bar (finishing a phrase), followed by "Play" later in the repeat, which should always be played
"Tacet first time"
"Play 2nd and 4th time"
"Last X only"
"On cue" (Of course there is no conductor to cue the computer...)

Of course, writing these instructions as system text works fine for most purposes, because I don't generally use Musescore to make recordings. Still, it's a reassurance to hit "play" and hear the computer interpret your music the way you intended it to be played.

If special notation is added, I would suggest adding only the simplest cases since it's impossible to predict how people will want to notate this concept.

IIRC, Sibelius interprets some of these special cases correctly. And if it doesn't, there's a manual control for it in the Inspector's "Play on pass" checkboxes.

Workaround: I don't think so, since we don't have unrolling of repeats yet. It becomes very impractical.

Priority: Certainly lower than, say, unrolling of repeats.

Problem to be solved

Get audio playback that better represents how a musician would play the music, in certain common scenarios of repeats.

Prior art

Sibelius 6 had this in Properties, and Sibelius 7 has the control in the Inspector. Screenshot from Sib 7.

Additional context

Previously reported in https://musescore.org/en/node/72791. Couldn't find a Github issue on it after looking at all open issues mentioning "repeat".

[Jolter]

An example of possible text notation: In this waltz theme, the tenor part should only play the countermelody on the second repeat.

[doncferguson]

+1. Working on big band scores. To minimize page turns and to keep things simple for the majority of players, would rather not unroll the repeats.

[joaoe]

Hi. I'm right now composing a piece with a big section of acoustic guitar, and there is voice that is sung only on the second repeat. Likewise, the last bar of the tutti where the ensemble plays a heavy chord in the end of the phrase coincides with the first bar of the guitar solo, and so that bar would have "1st time only".
This feature is essential not to force to unroll repeated sections.
Thank you.

[RedCoder09]

+1. This is super useful when orchestrating sequenced music, which often has layers that only come in after a repeat of the melody.

[ralphm]

This also goes for velocity. E.g. first time around piano, second time around fortississimo grandioso.

[ThePython10110]

TECHNICALLY this is possible but only in a very hacky way, involving creating an invisible measure (through measure properties), putting dynamics into it, and adding a frame with negative spacing to make things line up.

And it doesn't really work perfectly either, since all it does is set it to a very low dynamic the first time, which (at least in MuseScore 4, with its hardcoded dynamic values) is still audible, meaning the instrument needs to be turned down in the mixer, meaning all future dynamics need to be louder (probably meaning invisible dynamics every single time). It also adds a bit of visible space to the beginning of the following measure, and a 128th rest (assuming the invisible measure's length is set as short as possible), which could maybe be fixed using tempo changes... but yeah, basically it doesn't work very well and would be a great thing to have built-in.

EDIT: Just realized, this is not nearly as hard as I thought it was, as long as the repeated section is not at the beginning of the score. You just need to add a very quiet dynamic before the repeated section and a very loud dynamic right at the end (after any notes), then turn the instrument down in the mixer so the quiet dynamic is actually silent (which is why you need to use a very loud dynamic). I realize that in some cases, the need for a loud dynamic can make it sound bad...

[TiggerNZ]

I have seen innumerable scores where there are variations introduced in subsequent iterations, even if the dynamics do not vary (gradually increasing volume being a RELATIVE change being built from an initial state, as described by ThePython10110, before a repeated section).

Examples off the top of my head include descant parts and parts/instruments joining in for subsequent verses or the melody moving between parts/instruments, even certain verses being soloed by certain parts/instruments with others tacet.

Unwinding all of the repeated sections to render the audio is a fearsome prospect and a BIG invitation for error!

[Daniel-R15]

Came up on the forum, with reference to embedded dynamics for each verse of the lyrics:
https://musescore.org/en/node/364062