56303 DSP Eval Kit?

[Sean Costello]

Paul Perry wrote:

> The good thing abt the chip is that all the codec stuff is there.
> The silliest thing abt the DSP chip is that (contrary to advertising) it has
> very little
> in common with its 'predecessors', so far as code transportability is concerned.

Really? Does this mean that code like, say, the analog modelling code in
the Winter 1997 Computer Music Journal, which was written for a 56002,
won't run on the 56303?

A 5630x DIY project sounds pretty cool to me. I am slowly but surely
getting some computer music algorithms running successfully (I just
created a state-variable filter in C to be used as a Csound unit
generator, and the Bode-style IIR frequency shifter is next on the
list). I would gladly purchase a EVM and start coding, if other people
on the list are interested in such a project. I know NOTHING about
programming DSPs, but I know next to nothing about C, and I am doing OK
so far (no better way to learn). 

Does anyone know how the 56303 works with additional memory, i.e. can it
access enough memory to have lengthy delay lines? It would be cool to
have some more memory intensive techniques to play with, like delay
lines, reverbs, granular techniques, pitch transposition/timestretching,
FFT stuff, and so on.  The Nord and Waldorf synths that use the 56303
can't do any of the above techniques, due to lack of memory - probably
the main reason I haven't got that Nord Modular yet.

Sean Costello

P.S. Mar