First computer with real-time clock?

From: Paul Koning <pkoning_at_equallogic.com>
Date: Tue Aug 3 13:29:27 2004

>>>>> "Dwight" == Dwight K Elvey <dwight.elvey_at_amd.com> writes:

>> From: "Paul Koning" <pkoning_at_equallogic.com>
>> Consider GPS satellites, which have either or both built-in.

 Dwight> Hi I thought I'd mention that GPS satellites are very precise
 Dwight> but no longer accurate. The world standards like UTC are not
 Dwight> on the same second as used for GPS. I don't recall how far
 Dwight> things have changed but for political and other reasons, the
 Dwight> world time standards have been changing. GPS are still locked
 Dwight> to being relative to particular data and time. Every one else
 Dwight> has had a few leap seconds here and there. There is a chart
 Dwight> someplace on the web that shows how things are.

The fact that GPS time doesn't have leap seconds and UTC does, doesn't
make GPS time inaccurate. It just means there's a known adjustment
between the two, no different from what you do when you allow for
timezones.

I didn't realize that GPS time omits the leap seconds, but that makes
perfect sense -- if you tried to include them, it would just add a
completely unnecessary complication to the system.

By the way, leap seconds are NOT political. They are made (by the
BIH, International Time Bureau) when UCT -- which is basically time
reckoned by cesium clocks -- has drifted more than about 600 ms from
time as set by the earth's rotation. Since the rotation time of the
earth is not constant, that difference is subject to change, somewhat
unpredictably. So leap seconds are inserted on an "as needed" basis.

        paul
Received on Tue Aug 03 2004 - 13:29:27 BST