Re: Leap Seconds

["Marin David Condic" <marin.condic.auntie.spam@pacemicro.com>]

Bzzzzzzzzzttt! Wrong Answer! :-) I can neither change the hardware I'm
programming nor can I change the specification of the data being sent to me.
Its always nice to simply define your problems away, but sometimes you
can't. I could pile another layer of abstraction on top of it and hope to
surround it with an SEP field, but there are less drastic measures possible.

In this case, I'm not going to sweat it too much. Being off by a few seconds
is not going to kill the system and I am leaving a method open for the
operator to remotely download "tuning constants" which will include an
offset to time. If they really need to get the accuracy up to match the
incoming message, they can send me the number of seconds to adjust by.

It was a rather interesting question WRT how Ada keeps time and its nice to
know what is (not) required concerning leap-seconds. Its obviously not a
problem for most apps, but where it matters, I now know what is likely going
on under the hood.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"Wes Groleau" <wwgrol@ftw.rsc.raytheon.com> wrote in message
news:3B17B3B8.B0626F22@ftw.rsc.raytheon.com...
> With such a clear statement of the problem, the solution becomes obvious.
> You need to either get such a clock or design an implementation that
doesn't
> need one.  Which is what many of the posts were trying to do for you, sort
of.
>