Time signature changes cause incorrect measures from incorrect number of beats

Sorry, I can’t seem to reproduce this. That only happens for me if I have a double barline selected as the default barline in Notation Options, which obviously you don’t have here. There must be some other factor at play. If you can post a cutdown file with just a few bars that demonstrates that behavior, hopefully someone will be able to sleuth it out.

+1, completely agree!

1 Like

I eventually rooted this out, by finding it is a bug with Key Signature (plus perhaps also Time Signature at same location, and perhaps double-barline inserted at the same location). I re-used a prior project for a new project; the new project has key of C∆ instead of the prior G∆, and change of Time Signature at the same location. Deleting the Key Signature(s) and Time Signature somehow left the double-barline (there may have been a double-barline placed on top of this bar as well) when it should have been deleted; any attempt to remove or change the double-barline to a single-barline created the “q=” Signpost. The beginning of my new project sets the score to C∆, it remains in C∆ throughout. Finally in a much later edit, where I changed to a key other than C∆ at the start of the score, revealed these zombie Key Signatures, because Dorico placed natural signs there; I deleted these zombies, and the double-barline changed to a single-barline properly. My engraving now looks correct, but there is a bug here. Not the same bug as my original post, but something which at least appears related to the user, because of the resultant Signpost “q=…”.

If you add a barline from the right panel or from the popover, it is by definition a non-default barline, because you manually put it there. Given it would otherwise look like a default barline, a red signpost shows up in order to tell you that it’s a non-default barline, just like if you create a key signature with no flats or sharps.

This isn’t a bug.

To reinstate a default barline, you need to delete whatever it is that originally prevented the barline from being a default one, not manually add another new thing.

2 Likes

You did not comprehend the post detailing the bug.

Do you ever get the impression that (although some bugs do exist) it is tempting to call something a bug as a way of saying “Don’t blame me” for whatever went wrong?

1 Like

There’s much that you’ve not comprehended, starting with the fact that a red signpost means a hidden time signature (aka a non-standard barline, meaning one that’s not simply the result of an existing time signature). It’s nothing at all to do with bracket and barline changes.

Post an actual project and somebody will happily diagnose what’s going on.

1 Like


Interestingly enough, I managed to created a situation where’s there is both the hidden one and the normal one at the same time. Might it be a little bit more complicated?

I’d guess from the horizontal spacing that the second bar contains more than four beats. Again, without seeing the project I can’t be definitive.

You’re right! But then, why there is a singpost when there are more than 4 beats and no signpost if there are less?

PS : I’m not the OP, just to make sure

I’m pretty sure Dorico just treats these the same way as it treats pickup bars. Unfortunately, I think it also codes them that way in the XML data too. Finale’s “Check Region for Durations” plug-in would easily catch that in a native Finale file, but I recently tried using it on a Dorico XML and it didn’t catch the errors. Some sort of way to flag these for the user would be a huge timesaver when proofreading.

2 Likes

Hello - I was just tearing my hair out over a similar problem, arising from a meter change and resulting in a supposedly 2/4 measure with 8 quarter notes in it. Tried deleting the measure (which projected the problem onto the following measures) inserting new measures, inserting signposts, etc. WHAT FINALLY WORKED was using the rhythm popover to redefine the measure. Presto, everything back to normal. I still don’t know how it got started - this flow started life as an .xml file, but I’d been happily revising, changing meters, inserting measures, no problems. This one was STUBBORN and superblonde has all my sympathy. Was relieved not to have had to submit a new post - this one seems to have generated an unusual amount of friction!

This can only arise from two sources: 1) An error in the xml file; 2) User action.

There is no “rhythm popover”!

When dealing with time signatures in Dorico, there are a few important thing to remember :

  • Dorico will not change the total amount of musical time simply because you change a time signature (unless you are using Insert mode). All it does is redistribute the existing time into new bar lengths.
  • You can create a pickup bar at any point. This is a bar that purposely contains a different number of beats from the stated time signature (eg. you can create a 2/4 bar with 8 beats by using shift-M 2/4,8). Subsequent bars will all contain the normal 2 beats.
  • In doing the redistribution of time, Dorico will only do it until it reaches the next time signature (or manual barline change). This can result in one incomplete bar (and there has been much discussion as to whether this should be highlighted), unless Insert mode is ON (in which case the final bar will be padded with rests).

The genius of the Dorico approach is that you have complete freedom over changing time signatures safe in the knowledge that your music will never be lost, just re-notated according to the new time signature.

Hope this helps.

I meant the time signature popover.
(A rhythm popover would indeed be an interesting phenomenon.)
This all started when I wanted to change a 2/4 measure to 3/4. I’m fully aware of how to do this, including precautions, and have made many such changes with no surprises. I’m always on the lookout for those incomplete measures and
know how to fix them.
But this time, inputting 2/4 (after the new 3/4 measure) gave me a measure with 8 quarters in it… and no indication that the meter had changed, except the red description overhead. And I couldn’t figure out how to get rid of them. If I deleted the measure, the new measure was ‘rogue’ in the same way. Where did those extra beats come from? I think it must have been from the .xml. Earlier in the piece there is a ‘senza misura’ measure, in reality a 13/8 with a hidden time signature. I keep thinking those two rogue measures somehow got linked.

Anyway, inputting the correct meter with the popover solved my problem.
John

www.johnplantcomposer.com

I would check the initial xml for independent (local) time signatures and tidy them up first.
(but without seeing the file it is just a guess)

Just a guess (because I have made this “mistake” myself):
If at some point you made a selection with cmd/ctrl and the mouse, then moved to another location of your score and did another selection with the mouse and cmd/ctrl - the first selection might have still been active, resulting inadvertently applying a change to two places.