[time-nuts] NTP stratum 0

[Tim Shoppa]

"M. Warner Losh" <imp at bsdimp.com> wrote:
> : In any event, the noise from the frequency standards should be at
> : least a factor of ten smaller than the filtered noise limit of the
> : Internet, IMHO.
>
> It should be about a thousand times better...

While PPS with GPS receivers is easy and cheap and delivers that
level of accuracy, I think it's premature to claim that all refclocks
"should" be GPS based.

Diversity of refclocks has always been a principle of NTP and is a good
thing, even if they don't all deliver microsecond precision. Making
"NTP" synonymous with "GPS" is unfair to both.

Tim.