Yes, I'd read that. But it refers to voice numbers and I was not sure that the voice numbers were divided at the layer, just that half were controlled by one channel and half by another.
That's correct. It's true of both split and multi modes - half the voices are always allocated to each layer.
Particularly confusing to me is the last paragraph of the description "...Only Layer A of the selected program is loaded." and "Program Layer B data is completely ignored in Multi Mode."
I don't see that in either the version of the manual I was referencing or the version that's currently available from the Sequential support site. You might want to download the latest manual to check the current description.
I haven't used multi-mode much but I did test it out when I first got the instrument. I filed one bug (if the base channel was 16 then layer B was played on channel 2 rather than the expected channel 1.) I can't remember if it was fixed, but then it's something of an edge case.
Thanks for all you do to support the PX!
I'm happy to help where I can and when I have time. Obviously the instrument spoke to me, and I enjoy seeing others use it to its full potential.