well, then HS is only an improper MPE controller 
Edit: TBF your interpretation is obviously right.
As a matter of fact, one of ASM’s staff stepped out and said that HS is an MPE controller and in a way I can understand why that person was saying that. Nevertheless, when I had read that for the first time I was also struck by surprise because I would never have perceived the HS in this manner.
Usually an MPE controller will sacrifice some aspect of the dimensions and the Continuum for instance recommends against sending Note On Velocity and defaults to 127 as a fixed velocity for all notes.
In that sense the Hydrasynth outperforms the Continuum by providing reliable NoteOn/NoteOff velocities for the dimensions in the spec at the cost of some of the other dimensions.
This is semantic.
The Continuum simply replaces “velocity”, typically calculated by key-travel-over-time, with a pressure-over-time algorithm instead. This is because (and I’m paraphrasing here) the original vision for the EaganMatrix (and an ongoing debate in the MPE sphere) was to achieve expressivity without the need for static modulators like LFOs and envelopes. Never mind that the mechanism at play in the Continuum (for which the EaganMatrix was designed) differs from other MPE controllers, and certainly from that of your standard keyboard.
Regardless, the functionality of the Z-axis remains the same, and the Continuum still delivers all dimensions of touch, as would be expected of an MPE controller; albeit by way of a more elaborate set of calculations, post sensor.
So, in this case, Haken’s “recommendations” speak more to the method of synthesis being employed, rather than the MPE protocol itself.
Cheers!
As a sort of companion to this thread, i’ve created a thread specifically for MPE Software.