Do Logic Pro X Track Stacks play poorly with Smart Controls?

Question

New to Logic Pro X 10.0.4 here. I think I'm missing something, because I feel like Logic is messing with me when I try to use Smart Controls together with Track Stacks.

The way I can think to explain it is by example.

Setup:

1. Create a new Software Instrument track, and set it to patch "Female Chamber Choir". In Smart Controls for the track, set the choir to sing "Ah".
2. Create another new Software Instrument track, and against pick patch "Female Chamber Choir". This time, in Smart Controls, set the choir to sing "Uu".
3. Select both tracks, and then them into a summing track stack. The "super-track" that combines my two choir tracks will be called something like "Sum 1" or "Sum 2".

Now these things work as expected:

1. If I click on my first track and press keys on my MIDI keyboard, the choir sings "Ah", as expected.
2. If I click on my second track and press keys on my MIDI keyboard, the choir sings "Uu", as expected.

But here's what confuses me to death: Now I click the top-level track that's supposed to sum my two choir tracks (called "Sum 1" or "Sum 2" or whatever). My expectation is that I'll hear both "Ahs" and "Uhs" together, and that I can use the sub-volume controls to adjust the ratio between the two. But instead, I only hear "ahs". In fact, even if I mute the "ah" subtrack, such that all the input to the "Sum" track is coming from the "Uu" track, then still all I hear is "ahs".

My theory is that this has something to do with smart controls, because so far other uses of track stacks, when I don't use smart controls, are making sense to me.

Answer 1

Ok, I think the mystery is explained by "articulation IDs", a new concept added in Logic Pro X.

Here's what articulation IDs are about: Each MIDI note now can have (always has?) an "articulation ID" stored with it. Some ESX24 synth patches ignore this, but others, like several of the choir patches and several of the string patches, use them to decide which sample variant to play. For the Female Chamber Choir in particular, there are four different articulation IDs that the patch knows of, for "ah", "oo", "uu", and "mm" sounds, respectively. Each of these vowel sounds has its own totally independent set of samples, and the articulation ID tells logic which one to use.

Importantly, each MIDI note event can have only a single articulation ID associated with it.

As for the smart control connection: What the "Ah" "Uu" "Oo" and "Mm" buttons in the smart control do is change the default articulation ID for notes on that track. (It affects both MIDI Thru notes and recorded notes.) This, in turn, changes which samples the ESX24 sampler is going to play for each MIDI note. When you play MIDI thru data into that track, or when you record notes onto that track, each note will get marked with the track's default articulation ID.

Finally, here's what happens with track stacks, unless I've really confused myself:

• Say you select the "Sum 2" track and play MIDI thru notes with your MIDI keyboard (or, alternatively, say you record MIDI notes while the "Sum" track is selected).
• The "Sum" track needs to have a default articulation ID. Internet rumor has it (and this meets my experience so far) that logic looks at the lowest numbered default articulation used by its sub-tracks, and using that as the default articulation ID for the "Sum" track.
• In the case of the example here, it turns out that "ah" has a lower articulation ID (ID 1, I think) than "uu" (ID 2, I think), so logic picks the articulation ID for "ah" as its default articulation ID
• Now any notes coming in on the "Sum" track are going to be played with articulation ID 1, which means with "ah".
• Even though the default articulation ID for the "uu" track is not 1, that's just the default ID. The incoming MIDI data from the Sum track, however, gives each note a particular, default-overriding ID, and that ID, again, is 1, which, again, is for "ah".

To return to the original post but using these terms, the post thought that when MIDI data from from the "Sum" track got sent to each sub-track, each track's default articulation ID would determine which samples would get played. What actually happens, though, is that articulation IDs get set by the sum track before notes are passed to the sub-track, with the result that the default articulation ID doesn't have any effect. (Default only applies when notes don't already have an ID.)