Analogue-sequencer

>>Though did Chris Franke ratchet before the note, or after? 
What would be an example of a TD recording from which the answer could be deduced ?
Or is it a purely philosophical question ?<<

oo.... I thought I'd responded to this but it got lost in cyberspace...

"stratosfear", the first track. I reckon it'd be a bit of an arse but perfectly possible to set up on a moog with enough modules. you'd use the gate output from the preceding stage of the 960 to one-shot another sequencer running at a higher tempo, & sum the triggers. 

a similar effect can be achieved on the maq, incidentally, but it's even more of an arse to set up. the octopus can do it in strum mode, but again it relies on the strum being attached to the preceding note, which is less flexible.

anyway, the trill is clearly anticipating the beat. it can't be done the same way on a p3 because even though you can add repeats to the note before with the right pitch, there's one too many.
I can try to explain with a little four note sequence:

1 (* * *) 1 (* 3 4) 1 (* * *) 1 (* * *)

is what you're trying to achieve. the trill is before the third "1".

with the p3, you can either do:

1 (* * *) 1 (2 3 4) 1 (* * *) 1 (* * *)

or 

1 (* * *) 1 (* * *) 1 (2 3 *) 1 (* * *)

note that the trill is either too "busy" or comes after the "1".

d.