# The Naked Scientists Forum

### Author Topic: Comparing the accuarcy of two clocks.  (Read 3644 times)

#### phyti

• Jr. Member
• Posts: 37
##### Comparing the accuarcy of two clocks.
« on: 01/08/2006 02:57:34 »
We start with two copies of a common (lo tech-non atomic) clock.
Time zone changes are not relevant, and the clocks are compared to a bureau of standards clock.
The only rule, we can't adjust the clocks.
Since no clock keeps perfect time, we arbitrarily assume it loses time.
The time difference will grow to a maximum of 12 hours, then decrease to zero.
This may take years but the test requires a complete 12 hr cycle for accuracy.
The clock is correct, at the beginning,and end of the test, and has an average error of 6 hrs.
Clock 2 is stopped. For the same test period, it is correct twice a day, and has an average error of 6 hrs.
Conclusion, a stopped clock keeps better time than one that runs.
« Last Edit: 01/08/2006 03:59:33 by phyti »

#### another_someone

• Guest
##### Re: Comparing the accuarcy of two clocks.
« Reply #1 on: 01/08/2006 04:29:51 »
quote:
Originally posted by phyti

We start with two copies of a common (lo tech-non atomic) clock.
Time zone changes are not relevant, and the clocks are compared to a bureau of standards clock.
The only rule, we can't adjust the clocks.
Since no clock keeps perfect time, we arbitrarily assume it loses time.
The time difference will grow to a maximum of 12 hours, then decrease to zero.
This may take years but the test requires a complete 12 hr cycle for accuracy.
The clock is correct, at the beginning,and end of the test, and has an average error of 6 hrs.
Clock 2 is stopped. For the same test period, it is correct twice a day, and has an average error of 6 hrs.
Conclusion, a stopped clock keeps better time than one that runs.

Wrong conclusion.

Taking the parameters of your test (including that you are using rather antiquated 12 hour clocks ) – it may show that a stopped clock is correct more often then a clock that is very slightly askew – but it certainly does not show that a stopped clock will show the correct time more often than one that it running 100 times its rates speed (which would have the correct time about 100 times per 12 hours), or one running at 1000 times its rated speed (which would show the correct time about 1000 times per 12 hours).

In fact, a clock with zero error, but incorrectly set, will never be right.  A clock that has a 1% error will be correct once in every 100 12 hour cycles; a clock with a 100% error (i.e. running at either twice the nominal rate, or absolutely stationary) will be correct once in every 12 hour period; a clock with 1000% error (i.e. running 10 times the nominal rate) will be correct 10 times in every 12 hour cycle.

George
« Last Edit: 01/08/2006 05:16:23 by another_someone »

#### phyti

• Jr. Member
• Posts: 37
##### Re: Comparing the accuarcy of two clocks.
« Reply #2 on: 01/08/2006 06:04:53 »
We are only considering ordinary, common folks, everyday, run-a-the-mill clocks.
It is not a laboratory experiment using stobe lights, silver bullets, or large scale colliders.

#### sharkeyandgeorge

• Guest
##### Re: Comparing the accuarcy of two clocks.
« Reply #3 on: 01/08/2006 08:26:19 »
it was a poor premise phyti give it up.

J.B.S Haldane on the perforated eardrums which were a consequence of his pressure experiments "the drum generally heals up; and if a hole remains in it, although one is somewhat deaf, one can blow tobacco smoke out of  the ear in question, which is a social accomplishment".

#### JimBob

• Global Moderator
• Neilep Level Member
• Posts: 6564
• Thanked: 7 times
• Moderator
##### Re: Comparing the accuarcy of two clocks.
« Reply #4 on: 06/08/2006 16:10:14 »
I think the easiest way to ompare them would be to put them in a barnyard, wait and see which bird recognized dawn whenthe sun was at the horizon.

The mind is like a parachute. It works best when open.  -- A. Einstein

#### The Naked Scientists Forum

##### Re: Comparing the accuarcy of two clocks.
« Reply #4 on: 06/08/2006 16:10:14 »