Jack Ganssle's review of the Propeller

Paul Baker

Jack Ganssle, well known in the field of embedded systems, has written a review on the Propeller: http://embedded.com/columns/breakpoint/207000573·(It starts on the second page). Everything that's covered is well known by our users, but perhaps it will spark some new interest in·the Propeller.


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Paul Baker
Propeller Applications Engineer

Parallax, Inc.

Bob Lawrence (VE1RLL)

Cool! I get his Embedded Muse newsletter all the time.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Aka: CosmicBob

Devo

He makes this statement (emphasis mine):

Mr. Ganssle said...
A cog can read and write to the shared memory (well, they can only write to the 32 bytes that's not ROM), but cannot execute from that space. So programs are limited to 2 kbytes/cog, although it's possible to reload that memory at any time from shared memory. Two kilobytes might not sound like much, but many apps just don't need big programs.

I've not had a chance to play with a propeller yet (still on the UPS truck!), but this wasn't my understanding. Is it correct?

Devo

Gadgetman

It's true for a given value of true...

'Pure' assembly programs can't be longer han 496 Longs (slightly less than 2KB) and executes IN the COG,
then there's the Spin programs which are interpreted by an Assembly program which reads tokens from the HUB and 'executes' them. Then a single program CAN fill the entire HUB RAM.
And then there's the 'Large Memory Model (search for LMM here in the forums) which does some very tricky stuff with assembly, which I believe that not even Chip had thought of himself, and which loads assembly code instruction by instruction into a COG and executes it.
(I'm not entirely clear as to how it's done. I just know it works. )

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Don't visit my new website...

Devo

Aha. Thanks for the clarification Gadgetman!