2) Pitch bend and glide (portamento) are, in essence, the same thing
Well, yes and no. Yes, in the fact that both are bending notes. However, a synth handles this in very different ways. Portamento is essentially a voice parameter - ie, when you glide between notes, the synth uses the same voice and bends just that voice. It can do this independently for each voice.
Pitch bend is
not a voice parameter, it's a global parameter. It affects
all voices of an instrument at once. There is no such thing as pitch bending for specific voices (unless you do it some other way, such as vary the oscillator pitches, which
are voice parameters.)
This means that using pitch bend to achieve what you want is not trivial - you essentially want a unique pitch bend control for each note.
You then have two problems - how to transmit pitch bend to your instrument (which is a channel message, which means you can only have 16 independent pitch bend MIDI commands on a MIDI port); and how to get the instrument to respond to pitch bend individually for notes. Not only that, but with voices that have a longer release, pitch bend will also affect existing sounding voices that are in their release stage.
Doing it the way you want would require you to be able to transmit of 61+ MIDI channels, and have 61+ individual instruments, one for each note. Clearly, it's not an optimal solution. If you only want to handle a fixed number of notes, say 8 notes over one octave, then it's a little easier as you don't have quite so many to deal with. In this case, you could rig it up - setup an EXS24 for each note, and send your pitch bend messages across 8 MIDI channels to the respective instruments. However, this is restricting in the amount of notes you can play. I'm guesing you want to control multiple notes at once anyway (ie, all the C's in all octaves on one fader etc).
Fermusic's initial response that this is
easy was I fear a little misleading. While it's certainly possible, it's clunky, messy to set up, and quite frankly too complicated to bother with (imo).
So - what are the alternatives?
Logic does provide a microtuning facility for all it's instruments, but because these are preference settings, you can't really modify them by external controllers in real time, so it's not exactly what you are after. However, it will let you try out different scales.
Some other instruments might have the scale microtuning parameters automatable, and this could be a route to pursue. However, if you are stuck with the EXS24, the engine parameters you can "get at" are more limited.
Basically, what you want
could be done in Logic and the environment (one of the great things in Logic - it may not always be simple to do everything, but usually it is at least
possible) it's not the sort of thing that can be done satisfactorily in a couple of clicks.
If you are not that experienced in the environment, and still want to give it a go, I'm happy to give you some guidance in setting it up.
5) It cannot be denied that there is a limit to what the CPU can handle in terms of how many software instruments can be used at once
Sort of. In the case of the EXS24, and what fermusic was getting at, is that if you load an instrument into one EXS24 instance, that uses say 200megs of ram for the samples, loading that same instrument into successive instances of the EXS24 doesn't take up any more ram - the sample pool is shared between them.
And as CPU is only taken while notes are playing, to Logic it doesn't really matter whether one EXS24 is playing 8 voices, or 8 EXS24's are playing 1 voice each. The CPU hit is the same (although this isn't necessarily true of other instruments, which use varying voice optimisation schemes).
So in this case, runing 64 EXS24's, while they will take up a bit more RAM (because each plugin requires some RAM working overhead), it should be feasible to open up many multiple instances working on the same instrument.
