On Thu, 11 Oct 2007 09:52:53 -0700, Jim Lux wrote:
Frnak McKenney wrote:
--snip--
As for testing the clock's accuracy, you're right about needing a
second source ("Qui custodiet ipsos custodes?" or something like
that? grin). On the other hand, as long as the digits are
flashing by, I'm happy to "just trust them".

One needs three clocks.. if you have only two, you only know that they
differ. If you have three, you can detect the failed clock, because the
other two read the same time.

Well, you have a reasonable _presumption_, anyway. grin

In _theory_, one could have two clocks in error and one on time. Or
even all three out of step with (say) the NIST's clock. But
two-out-of-three would be the place to put your money. grin

Of course, if the different clocks have different accuracies or
reliabilities, that's another story.

For a more detailed treatment of such things, you might want to check
out the "Byzantine General" problem.

Faint recollections... "unreliable communication on the
battlefield" category?

Based on the feedback from other posters, it's likely a consequence
of 10MHz propagation. A VLF RF signal like 60KHz reportedly does a
much better job of getting a readable signal to a wide area.

Although in my house (southern California), the 60 kHz signal seems to
fade in and out on a daily basis.

I can believe it. I've seen those LCD "clock and weather" stations
take days to synchronize the time (I assume they're using WWVB).
OTOH, I'm hardly one to talk, since I've let the MAC-II sit
liostening for months at a time without successfully locking-in on
WWV.

Frank
--
Hanlon's Razor: ÿNever attribute to malice that which can
be adequately explained by stupidity.ÿ
--
Frank McKenney, McKenney Associates
Richmond, Virginia / (804) 320-4887
Munged E-mail: frank uscore mckenney ayut minds pring dawt cahm (y'all)