V 3.5.12 Bug? Inconsistent bar numbering - related to independent time sig change

I’ve come across something that caused much confusion at band practice this evening. The attached cut-down project shows the problem. The pickup bar in the last flow is counted as 0 in the flute, oboe and bass clarinet parts, but in the others, which all have independent time signatures, it is counted as 1. Also the bar numbers are not displayed in the system track for those part layouts.

I can only assume it has something to do with the presence or otherwise of the independent time signatures, but not sure why this would be, or what I can do about it other than set bar number changes manually on the relevant parts.

Holst Second Suite cut.dorico (520.6 KB)

Sorry, James, but looking at your project, in the fourth flow (Fantasia on the “Dargason”) I’m not seeing any independent time signatures – what am I missing?

In any case, you can select the independent time signature itself and create a bar number change that will apply just to that time signature, so you can set things right that way.

I maybe should have said “independent time signature changes”. Bar 58 in the euphonium, 75 in clarinet 1, and returning to orig time signature in 88-89.

In the cut version they are the only two parts affected by this problem, but in my complete version two other parts have time signature changes later on and they are also affected. The bar numbering goes out right from the start of the flow, which is what I find odd. I realise it’s simple to fix, but this can’t be intended behaviour surely?

Right, I see what you mean now, thanks for the additional clarification. It comes under the realm of undefined behaviour, I’d say: when an instrument uses an independent time signature, it starts to derive its barring (including bar numbering) information from another source that ignores the entries in the global list. I agree that this is something that should ideally work differently, and when we return to work on polymeter we’ll try to square this away.