Working on a concerto and now the soloist has an extended cadenza. It is pretty freely metered; mostly 4/4 but sometimes some 6/4 and sometimes some really long cadenza-ish measures. And the composer didn’t bother to indicate meter anyway, although there are logical bar divisions.
So I thought to write the entire cadenza in open meter and put in the explicit bar lines where I want them. I hadn’t realized, but that actually creates a new open meter signpost at every bar, so now the problem is in the instrumental parts – rather than having a single multibar rest for the cadenza, every bar is a whole rest.
Is there a way to consolidate those rests into a multibar rest for the parts? Or is there another way I should do the cadenza so that it is freely metered but the parts show an extended multibar rest? I’ve tried putting the solo part in open meter and the remaining parts still in 4/4, but whenever I put a bar line into the solo part, the parts still split at that point (with a new hidden signpost meter of 4/4).
Part of the trick here is using local time signatures and barlines in your solo part – which you can add by closing the popover with Alt/Opt+Enter instead of just Enter.
Here’s an explanation of one approach:
If you scroll a little further down in that thread, you’ll find a slightly different approach that preserves proper playback, if that’s important to you.
Thanks for the reply. Putting in local time signatures doesn’t really work, because if the time signatures are not the same, the measures don’t line up in the score. The only way to get the measures to line up in the score is if the meter is the same, which is why I was hoping an open time signature would work.
Seems like I’ll have to fudge the soloist measures by setting up things like 24:4 tuplets for the large bars and hiding the tuplets (and changing the spacing in engrave mode).
I did that by putting a local 4/4 time sig in the top staff and local dashed barlines as needed. To get 6 quarters in the second bar and 3 in the third, I used 6:4 and 3:4 tuplets and hid the bracket and number. (If you have a “really long cadenza-ish” measure, you can use whatever kind of tuplet you need – like your 24:4. I don’t think you should need to adjust spacing.)
I assume you’ll probably have a rehearsal number or something after the cadenza, which will break the multibar rests in the parts, but if not, you can apply an explicit global single barline there (which is what I did).
Edit: The local meter in the top staff could just as easily be open, if that’s what you prefer.
It was a good point about the single vs. multi-bar rest. I don’t actually care whic it is as long as the part makes sense, but also the measure numbers need to match. But doing a single open bar in the parts and multiple local bars in the soloist works well, and then I can add a measure number change when the orchestra re-enters. Though that single bar in the parts is by default quite large and needs to be manually resized to something reasonable, and there are some complications around that if I want to put in tempo changes for the sololist. Still, a good workaround for now; thanks for the input.
The single bar in the parts should be just the size of a regular 4/4 bar. It just looks big in the score because it needs to accommodate all of the music in the cadenza.
If I understand your latest example: you knew that the cadenza was 20 beats so you entered that as the initial tuplet for the soloist, put in the first four notes, then put in a local open meter, then put in the remaining notes and local bar lines. That ends up with a regular single bar rest for the piano and seems a reasonable approach for a simple case (and has a normal sized bar for the cadenza in the part, since it has a 4/4 time signature).
In my case, with a lengthy cadenza of 40+ bars anywhere from 4 to 74 beats, this seems a somewhat fragile and hard approach. The initial tuplet would be 515:8 (if I counted correctly…); just for fun I tried putting that in a new document and it seemed pretty unwieldy to work with (particularly as it is a piano cadenza with multiple voices). Not sure I could follow the same approach you did, but knowing in advance that initial tuplet size seems to be crucial to get the ending bars to line up.
What I had found to work well was to set everything to open meter, then set the soloist to local open meter, and then enter the notes with local bar lines for the soloist. (Another advantage of this: the instruments that need to cue the last few bars of the cadenza can also use a local open meter, put in the last few local bar lines, and enter a cue from the soloist.)
But that open meter measure that ends up in the parts: that’s what ends up with a huge calculated size. Here’s a quick sample where the cadenza bar in the first system had 50 beats in the soloist part; you can see the cadenza bar on the first line is way oversized, particularly compared to the actual empty measure on the second system:
