Tracy Smith wrote:
> I'm sure that Wagner, like always, has good
> intentions, but don't let his absolutisms fool you.
>
> The subject you wish to learn more about is: phase
> accumulators. Do some home work and search about this
> on the web. Others have commented (quite succinctly)
> on this list in the past. I remember posts from Paul
> (not sure which one) and Mike Keitz that were both
> very good. So perhaps the archives would be a good
> place to begin.
[snip]

If you have jitter, even when the total pulses per second is correct,
you have an imprecise wave form, that serves only as the average along a
period of time, nothing else, even that several frequency counters (HP
5328, 5334,5,etc) would say it is a correct frequency (average in time),
but try to measure pulse period.

Of course it can be used in several application when pulse period
accuracy is not a request. I also think that a frequency generator based
on a 8MHz PIC will not be the "important" time base in a bench, is it?
so anything will works for the purpose.

Price by price I would go for a commercial programmable wave form
generator and then use the PIC to control it.  I understand that it
could be funny to use a hammer to set a bolt, but it is not the right
tool, you will have *limitations*.

Try several "prime" numbers as frequency to be generated from a 8MHz
driven unit, and after a while even your calculator digits will be
jittering... :)

You will not want a microcontroller being driven by a jittering
programmable oscillator, if you need clock stability and precision.
Don't let miraculous solutions fool you.

Remember the "Magician Law":  The effect of a magic act is directly
proportional to how much fool is the audience, so, when you see a great
magician producing a great magic act, he is only exposing the big fool
in you.

Wagner