Thanks, Sandy. That makes sense.
A few years ago a colleague was using a website with the accurate time
displayed, supposedly. But it was a good 5 seconds from my watch, which I have
kept within a second for the last 35 years. It has become a habit, but my job
required that sort of accuracy: delivering weapons on a target had to be done
on time as there had to be a predictable interval between aircraft attacking
the same target.
It's not as important now, but I do like to be 'on time' nonetheless :-)
Chris
On 14 Aug 14, at 06:45, Sandy Harris <sandyinchina@xxxxxxxxx> wrote:
> NTP (network time protocol) deals with the latency.
> http://tools.ietf.org/html/rfc5905
>
> The basic idea is send a query to a server, record the time you wait,
> get a response. Say it takes 20 msec from query to response; add
> 20/2=10 msec to received time and use the result. If you need better
> accuracy, use several queries, possibly to different servers, and
> average the results.
>
> This is definitely not perfect, but it gives useful results -- much
> better than 1 s accuracy -- in most cases. If you need something
> better, like millisecond accuracy, a Rasberry Pi with a GPS attachment
> can provide an NTP server on your network for not a lot of money.
> --
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|