[time-nuts] An (unknown?) nasty feature of the DDS principle for time nuts applications

[Poul-Henning Kamp]

In message <E34BA9BA1D7340AD88EF19CDD971F3D8 at athlon>, "Ulrich Bangert" writes:

>However for very precise timing a DDS may simply be unsuited. 

No, it just has to be correctly designed.

For integral Hz resolution output, you want to feed it a frequency
with us 2^N(*10^M) where N is equal to the width of the accumulator.

However, don't overlook the value of a small frequency offset, which,
if used correctly, can be used as a 'vernier' to discover minute
unlinearities in timing circuits.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.