Re: [gpsd-dev] Clarifications needed for the time-service HOWTO

[Hal Murray]

> > .Summary of typical accuracy
> > | Stratum 2 NTP time  |  ±100mSec
> > | Stratum 3 NTP time  |  ±200mSec

> These figures are from the End Run Technology page on Stratum 1 (cited).  I
> have been unable to find another public source that is willing to commit
> itself to an accuracy bound.  If you know of one or more, by all means point
> us at them; they'll go in the references. 

I'd call that BS.  The stratum is much less important than other things.  The 
primary consideration is how well the servers are chosen, either by luck or 
skill.

You might get some data from the pool.  Beware, their monitoring is all run 
from one host so it will get a bad opinion of servers that are far away (or 
unlucky).


>> With this configuration, ntpd will read the timestamp posted by gpsd
>> every 64 seconds (16 non-root) and send it to unit 0. 

Where does that "16 non-root" come from?

The word "poll" is confusing in this context.  ntpd calls/polls the SHM 
driver every second.  If a sample of data is available, it gets put into a 
buffer/fifo.  Every polling-interval, ntpd looks in that fifo, discards 
outliers, averages the rest, and feeds that answer to the main timekeeping 
logic.  The polling-interval defaults to 6 (64 seconds) but can be adjusted 
by setting minpoll/maxpoll.



-- 
These are my opinions.  I hate spam.