The theoretical boundaries are limited only by 1) the amount of memory available on the flash ROM to store the native operating system including all the routines and instructions, and 2) the processing power of the CPU to attempt to carry it all out in realtime.
I guess Access try to maximise use out of both (Flash ROM memory and CPU).
A given CPU can only process instructions at a certain speed. If the CPU is being utilised heavily and you add another LFO on top of what it's currently processing, requiring more instructions to be processed in realtime than it can actually handle, it might be enough to overload the CPU and you end up with glitchy audio or drop-outs.
Similarly, if you have filled up the Flash ROM with routines and instructions and the given amount of memory storage is nearly full, you wont be able to add any new software functionality to the synth (in other words you wont be able to teach it new tricks).
In other words, when Access release a new synth, the Flash ROM is not full, meaning that new routines and instructions can still be added at a later time. And the CPU they choose to use also has room to spare, in that it is still able to carry out more instructions per second than is required for the original OS (when the synth is originally released). A bit like a car engine that is capable of doing 130mph, but only 100mph is required of it at the beginning.
Access may also fine-tune and optimise their code over time, meaning that it can use less filespace on the Flash ROM, giving them additional space for new updates.
__________________
PS > And another thing! Will the Ti|3 have user customisable/importable wavetables?  A ribbon-controller or XY-Pad might be nice, too, please! Thanks!
|