hi all,

i did some more research on the cross-compilation thing and realized it is better to do this in pf. it has all the required machinery to make it work, can run standalone and can be hosted by pd.

mole is currently too much of a liability to use as a basis for building the pd<->microcontroller thing. once the project works it should be possible to make it run in mole, which is much smaller because it doesn't have all this packet crap, until then i will concentrate on pf and send all remarks and comments to the pf-list.

i've just sent a large email to the pf-list explaining the roadmap for hooking up virtual and real machines to pf. check:

http://zwizwa.fartit.com/pipermail/pf-list

as a test case i am porting the scaf virtual machine (simple cellular automaton forth) to pf. this works, not counting some small bugs. still, it's 'virtual' :)

the only thing in the way at the moment to make this microcontroller thing work is the actual hardware and the bootstrap interpreter, which could be the 3-instruction forth.

cheers

tom