[time-nuts] Notes on tight-PLL performance versus TSC 5120A
warrensjmail-one at yahoo.com
Thu Jun 3 16:54:43 UTC 2010
Steve Rooke posted:
A whole old bunch of stuff, this is all I found really wrong:
>OK, we need to run the test for a longer period, ...
we can never prove it true so we try to look at ways of proving it false.
The important words there are "proving it false".
That tester probable has more hours of compare reading on it now than most
It don't really need no more longer runs to show what it is going to
Stop this insane proving it either true or false, So far it is 25 timers
been true and once has been false.
But that false time was when John had it connect backward or something like
that, so we can list that one under operator error.
so 25 timers good, 0 bad, one operator error
As Bruce has pointed out I'm not real good at statistics, but that sounds
like it is about a 25 to 0 chance that the next 25 test will also be OK
Sounds like a saver bet to me to guess 'True' , If one is into betting the
odds and not King worship.
God save the king
More information about the time-nuts