On Jul 7, 2005, at 1:30 AM, Wouter van Ooijen wrote:

>> So what will you suggest one to learn, dsPIC or AVR or even ARM if one
>> wants to upgrade his or her programming skills on medium end MCU
>> (higher end than the PIC12/16 or even PIC18)?
>
> IMHO the step above the PIC/AVR should not be the 30F or a high-end AVR
> but an ARM or another high-end processor (I am not familiar with 
> others,
> but they probably exist).
>
I don't think that the AVR or dsPIC are sufficiently "higher" than
the PICs to be a next step "up."  ARM probably qualifies.  Also
68k, coldfire, powerpc, maybe Z80.  And x86 embedded variants.  Perhaps
MSP430 (high end), although that lacks some in the cpu power department.
You're definitely into the realm of von Neuman architectures (shared 
code
and data address space), unless you want to look at midrange DSPs...
It depends which part of 'higher end' you're interested in, mostly.
And you also enter the realm where board design gets replace by SBC
buying decisions; there are usable platforms in the form of palmtops,
game machines, off-the-shelf routers that happen to run linux,
Whole PC-architecture things no bigger than an SODIMM,
cheap last-generation desktops, and laptops.  Lots of choices.

One of the reasons PICs are fun is that they tend to fit problems where
there is still opportunity to exercise a fair amount of creativity.  
Move
much 'higher', and you flirt with becoming a slave of the performance
that such a machine is expected to have, and/or someone else's operating
system or architecture.

But I think the next step up would involve learning "unix programming"
in C.  Many architectures run something that looks like unix, and many
others provide and/or emulate the unix ("posix") libraries...

BillW
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist