Difference between Primary and Secondary volume (dynamic) control in Expression Maps?

I saw the tutorial video and it mentions the new feature in Dorico 3, now you can set a secondary volume control for expression maps for virtual instruments that use two. In my case I have a string VSTi that uses CC1 (Mod Wheel) to set the overall dynamic and CC11 (Expression) which is used for manual minimal changes to the overall volume in order to produce more expressive and realistic phrasing; Velocity is completely ignored. So my question is, given that in my case CC1 is the ‘‘important one’’ and CC11 is used AFTER everything else to create a better phrasing, which one should be set as the primary dynamic control and which as the secondary one, and why? How does Dorico 3 use the secondary dynamic different to the primary one when both are different CCs?

I looked at the manual and the Expression Maps section is literally copy pasted from Dorico 2, which doesn’t make sense because in Dorico 3 the Expression Maps window got some changes, so it doesn’t mention anything of this and I believe it should be addressed. I looked at the Dorico 3 Version History PDF and it doesn’t say how does Dorico use these parameters when both Primary and Secondary volume control are set as different CCs either (it only mentions if CC1 is set as primary volume/dynamic control and Velocity as Secondary).

Even in the example given by the Version History PDF, it doesn’t exactly explain how does Dorico utilize that information… Does Dorico generate different MIDI data for the Secondary volume control or does it just copy the MIDI data from the Primary one?

Also some playback questions: Does Dorico try to create some phrasing (intensity changes) for playback? I know dynamic markings and hairpins are translated to the Primary Volume Control set in the Expression Map, be it Velocity or a CC, but does Dorico also generate minimal variations in order to create phrasing? If yes, can that data be sent to the Secondary Volume control while only sending the Dynamic markings to the Primary Volume Control, following the rationale used by my VSTi when using CC1 and CC11?

I just had a brief look at this - dumping MIDI into Cubase’s monitors - and it seems that the secondary volume is just a copy of the primary. Also, it seems that the key velocity setting for the actual key switch notes still is ignored - they all come out as 100 in Cubase, which means there’s still no efficient way of triggering variants of the same technique (such as different attack types for legato e.g. - Berlin strings come to mind… without resorting to extra programming and workarounds… )

In due course we anticipate you being able to set a different curve for the primary and secondary volume controls, so that at the same dynamic they will output different values. At the moment, everything follows the power curve defined on the Dynamics page of Playback Options.

I see, so it’s better to just ignore the Secondary volume control in my case, setting CC1 as Primary and manually drawing a phrasing curve on CC11 on the Play mode on each track that requires it. Thanks for finding out.

there are a number of different ways of looking at this, using the VSL strings as an example. I would say that for expressive results in general you will want to use Vel. XF to make use of the different velocity layers but that of course overrides the default volume. So why not use velocity as default primary controller so you can see the dynamics in the score but duplicate to CC2, or whatever, for the Vel. XF as secondary. Add CC7 as a genuine independent secondary volume controller for smooth hairpins etc and it seems to me you’re starting to get something quite nice. And it’s one way of the Expression Map secondary controller being genuinely useful. But I stress I’m still in relatively early stages of looking at this and there may be drawbacks to this approach.

Main point is that you get a proper overview of all dynamic controllers at work in PLAY mode (although the elsewhere suggested addition of being able to view more than one controller at once would certainly be a welcome addition) which has only been possible in a DAW to date.

UPDATE
looks like I’m coming round to velocity as the main volume control if you’re using Vel. XF for the simple reason that the instrument will respond to the velocity lane levels which are set in the score. To modify, use the Velocity XF slider which changes the timbre (CC11 or 12 are normal but it doesn’t matter). The controllers are set in VSL, obviously – in Dorico, you only change the levels (and obviously the techniques) and I’m not sure that it matters what the Expression Map settings are. Am I missing something here?