[ntp:questions] NTP vs RADclock?

unruh unruh at invalid.ca
Sun Jun 3 18:56:29 UTC 2012


On 2012-06-03, skillzero at gmail.com <skillzero at gmail.com> wrote:
> Has there been any independent comparison of NTP vs RADclock [1]?
> Information on the RADclock site seem to indicate it performs pretty
> well, but I haven't seen any analysis except from the RADclock
> authors.
>
> I'm using NTP today for synchronization between devices on a LAN to
> each other of an internal clock separate from the system's normal wall
> clock (which uses the system's NTP). So I have flexibility with update
> intervals (1-3 seconds acceptable in my case) and other parameters.
>
> [1] <http://www.synclab.org/radclock/>

Not much info I could find there, but it sounds like they separately
track the rate and the offset of the clock counter, instead of having a
clock which just gives you the time. Ie, the rate clock could be way out
in time, but gives good Delta t, while the rate of the other clock could
wander (as it does in ntp) to try to keep the time on course. 
Not sure that is what you want. 
Sounds like you want your internal clocks to be such that TimeA-TimeB is
kept small, but you do not really care if TimeA is the right time, 
Of course the best way might be to sync them both to say a GPS clock
separately, which you can do to the microsecond level, rather than use
the net to try to sync them to each other ( which may only be good to
milliseconds)

If timesync is what is important, then chrony does a bit better than
ntpd does. 



More information about the questions mailing list