Hi Peter: I think the beginning of the idea goes back to pen plotters where you may need to draw a line where the end points are not on the grid of integers of possible pen positions. But the remainder may be off by 8 bits. For more on Bresenham's algorithm see: http://www.pacificsites.com/~brooke/PClock.shtml#BA The problem with it is that the jitter can be as large as the register.
It's possible to get the same effect but with a jitter of only one clock period. To do this you divide the input frequency by N for X cycles then divide it by M for Y cycles. For example you can divide any oscillator frequency down to 1 PPS and the jitter in the 1 PPS period will be only 1 uC clock period. You don't need to use dual modulus counting if the input frequency can be factored into numbers that are smaller than the register size of the micro controller. Have Fun, Brooke Clarke w/Java http://www.PRC68.com w/o Java http://www.pacificsites.com/~brooke/PRC68COM.shtml http://www.precisionclock.com Peter Vince wrote: >David Forbes recently mentioned the Wikipedia item on dual-modulus >prescalers: > > http://en.wikipedia.org/wiki/Dual-modulus_prescaler > >I have looked at that article, and either my brain has slipped out of >gear, or there are some errors in both the diagram and the words:-( I >think I can see the sort of thing they are trying to achieve, but was >hoping someone here could provide a clearer explanation please? > >Thanks, Peter > > > >_______________________________________________ >time-nuts mailing list >[email protected] >https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > > > _______________________________________________ time-nuts mailing list [email protected] https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
