Re: PIC clock.

Richard Seriani wrote:
"Anthony Fremont" <nobody@xxxxxxxxxxx> wrote in message

I disagree. It's an obsolete part that is vastly inferior to the
_cheaper_ alternatives. To convert f84 code to run on a 628 or 88
consists of these three steps:
__CONFIG adjustment
Account for start of user RAM being moved to a slightly higher
address Initialize CMCON on the 628 (ADCON on the 88)

That's it. That's all you have to do. It's trivial compared to
effort of trying to make sense of banking.



ANSEL (analog select), so we don't try to drive digital signals into
an analog input, instead of ADCON?

Oops sorry about that. I didnt' verify it, I was just spouting from memory.
;-) One other "tiny" detail is that CMCON changed banks between the 628 and
the 88.

I, too, have used the 16f88 in place of the 16f84a, but it seems most
courses (high school and university) start out teaching the 84a.
Probably because there is so much information available. Of course,
as long as the required results are demonstrated, I haven't yet found
a professor who won't allow use of a PIC which has more features (the
16f88 and 16f87x seem to be quite popular). In fact, it is sometimes
encouraged in order to disuade students from totally plagiarizing
code from the internet.
Which chip, manufacturer, or chip family to teach at the entry level
is, in my opinion, less important than getting the students
interested in the idea of being able to make the MCU do something -
anything - they program it to. Those who want to pursue it will have
the spark ignited, visit sites and newsgroups that espouse various
technologies, and make their own decisions about which component,
language, and development environment best fits their current
project. Those who only want a grade with little effort or real
desire to learn, won't - no matter which chip was used in the course.
Thanks for reading.

Richard


.

Quantcast