On Wed, 19 May 1999, Lance Allen wrote:

> Has anyone here used the DDS method to generate sinewaves with a PIC?
> We used a 16c74 running at 8MHz to generate 1Hz to 2000Hz in 1 Hz
> steps. Great for the original application but now we want to ramp up
> to 20kHz in 1 Hz steps but we just cant squeeze enough for 20kHz out
> of the code (now a 16MHz xtal). Has anyone done this?  Some clues for
> tight code?... please.


Have you seen:

http://www.brouhaha.com/~eric/pic/sine.html

Or

http://www.interstice.com/~sdattalo/technical/software/pic/picsine.html

The former implementation (Eric Smith's) is faster, while the latter
(mine) has more frequency resolution. A DDS implementation with my routine
would require at most 80 cycles (and I know a way to reduce it by 20) and
would easily generate a 20kHz sine wave with milli-Hz (or perhaps ten's
of) resolution. Eric's routine would probably allow generate a 150kHz sine
wave, but with less resolution (frequency that is). Your Choice...

Scott