[BUG] Stem direction of tied notes changing arbitrarily when inputing a higher voice.

The lower voice in this staff is 1st voice, stems up, with forced stem direction down. Here are some pictures of the issue. Ill attach the Dorico project in question so you (Daniel), can try imputing the note yourself to replicate the bug.
Captura de pantalla 2017-12-11 a las 12.21.30.png
Captura de pantalla 2017-12-11 a las 12.21.46.png
Just as a side note, i’ve encountered this bug in many other projects now, i’ve decided to write this post as it’s getting somewhat annoying.
Tied notes stem switching.zip (518 KB)

Why don’t you use Downstem Voice 1 for the lower voice, rather than forcing stems down in an Upstem voice? It’s what it’s there for…

That’s what I ended up doing, it’s still a bug however. It works for this situation, but when transcribing more complex music (Bach fugue), with more voices per stave, it becomes apparent this is a relevant bug.

My guess here would be that when there are three or more voices in a bar, Dorico’s only recourse to make things “correct” is to honor the default voice directions. With only two voices there is more flexibility to force things one way or the other, but once you exceed that, then it makes sense that the voices would be forced to the “normal” position because it becomes too complex to auto-adjust with a satisfactory result.

That’s not the case here anyway. There are only two voices at all times in the upper staff.

I don’t agree that this is a bug. You should be using voices with opposite stem directions.

I think I misread the first pic and I hadn’t opened the file; I assumed there was a 1st downstem voice since there was a second upstem. Either way, I have to agree with Daniel… I work on contrapuntal organ music with up to five lines of counterpoint per stave and I’ve quite literally never had a problem with stem directions. The good news is that now that there are new filtering features, it is really, really easy to get the desired result.

Just select the upstem voice #1 starting in MM36 via filters, and then use the voices->swap into downstem voice 1 command. Takes 30 seconds.

Well, if the feature “force stem up” is used, and it doesn’t work (or in this example, it does only work on the first tied note), how is that not a bug? I get that I can change the result by changing the whole voice, but that isn’t the point I’m trying to make.

You don’t have to force the whole voice: in Engrave mode you can select the tied-to note and choose Edit > Stem > Force Stem Down.