time-nuts@lists.febo.com

Discussion of precise time and frequency measurement

View all threads

Algorithm in use by chrony and OpenNTPD

WL
Watson Ladd
Thu, Jan 30, 2020 6:30 AM

Dear fellow nuts,

I'm trying to find any sources on the clock discipline and source filtering
algorithm that chrony and OpenNTPD use. It's different from the one ntpd
uses that Mills described, which is much more like a PLL one might build.

Sincerely,
Watson

Dear fellow nuts, I'm trying to find any sources on the clock discipline and source filtering algorithm that chrony and OpenNTPD use. It's different from the one ntpd uses that Mills described, which is much more like a PLL one might build. Sincerely, Watson
MD
Magnus Danielson
Thu, Jan 30, 2020 9:41 AM

Hi Watson,

On 2020-01-30 07:30, Watson Ladd wrote:

Dear fellow nuts,

I'm trying to find any sources on the clock discipline and source filtering
algorithm that chrony and OpenNTPD use. It's different from the one ntpd
uses that Mills described, which is much more like a PLL one might build.

You should start off by reading the many small articles that
Poul-Henning has written, they are here:

http://phk.freebsd.dk/time/

What he does is in his own words, non-magical, and I agree.

Cheers,
Magnus

Hi Watson, On 2020-01-30 07:30, Watson Ladd wrote: > Dear fellow nuts, > > I'm trying to find any sources on the clock discipline and source filtering > algorithm that chrony and OpenNTPD use. It's different from the one ntpd > uses that Mills described, which is much more like a PLL one might build. You should start off by reading the many small articles that Poul-Henning has written, they are here: http://phk.freebsd.dk/time/ What he does is in his own words, non-magical, and I agree. Cheers, Magnus