Re: ATM Synching to atomic clock via SNTP...

[Mark Alsing <malsing@borg.ae.usr.com>]

Bill,

Thanks for the info on The Sky. This is, in part,
what I was looking for. It seems there is a way to
sync to an external time standard and thats good.

Now, regarding your comment about varying accuracy
of PC (or MAC) clocks. Since these systems are not
meant to be operated in the harsh environments of
observatories thier clock circuitry can drift
pretty bad over just a short time due to thermal
and humidity factors. I have seen some equipment
loose or gain as much as 5 minutes an hour in a
thermal test chamber.

Over the course of an observing session lasting
several hours I wonder if you would have to keep
dialing the USNO to re-sync? Thats long distance
for me and most people. My ISP is my company and
I dial an 800 (toll free) number for my access.
I could simply stay connected and re-sync my clock
every 60 seconds if necessary with a time client.

Now comes the important question that I fear only
the vendor can answer. Is The Sky or any other
astronomy software "smart enough" to re-read the
system clock (hardware) periodicaly to make sure
it is in sync with the system (which is in turn
synced to a stratum 1 timing source) or does it
simply read it once on start-up or after dialing
the USNO and maintain timing in software after that?

I guess one way to find out would be to run one
of these programs then change the system time by
an astronomicaly noticable amount and see if it 
notices the change. This change in time is effectively
what would be happening everytime the systems clock
was reset by the time client.

Thanks-Clear Skies!
Mark Alsing
Network Systems Engineering
3Com/US Robotics, Inc.

On Fri, 4 Dec 1998 UC007@aol.com wrote:

> 
> I am not sure that I'm using the SNTP client, but I correct my clock regularly
> using The Sky level IV software. When I tell it to, The Sky dials
> 1-303-494-4774 and connects to time signals put out by the U. S. Naval
> Observatory Atomic Clock. It works like a charm. What it does is correct my
> computer clock which inturn corrects my LST display and corrects time
> functions in The Sky which relys on the computer clock time base. Whether the
> computer clock then maintains an accurate time base over, say, an observing
> period of several hours would have to be ascertained for each individual
> computer depending on the accuracy required, and a correction applied to
> compensate for accumulating error.
> Bill
>