[ntp:questions] Any chance of getting bugs 2164 and 1577 moving?
unruh
unruh at invalid.ca
Thu Mar 22 20:21:23 UTC 2012
On 2012-03-22, David J Taylor <david-taylor at blueyonder.co.uk.invalid> wrote:
> "unruh" <unruh at invalid.ca> wrote in message
> news:uNJar.12581$QC3.8535 at newsfe16.iad...
> []
>>> Most likely I would be looking at a histogram of the reported offsets,
>>> and
>>> see whether it was gaussian, flat, or whatever, and how wide. I might
>>> learn something from that.
>>
>> No. Not if it is just noise.
>
> .. but until I see I won't know.
And if you saw it, how would you know?
>
> []
>> precision is not accuracy.
>
> and where did I say it was?
>
>> In science we teach students not to report unwarranted precision-- the
>> precision should reflect the accuracy of the measurements. We keep
>> getting measurements to the mm and reported precision to angstoms
>> because that was what the calculator spit out.
>
> I hope you teach error estimation as well.
>
>> I am not averse to reporting with a precion maybe up to a factor of 10
>> better than the accuracy, but any more is just silly and misleading (as
>> you are demonstrating in believing that a greater precision would convey
>> some extra information.
>
> Should you read what I wrote, including the bug report, perhaps you would
> see that I was quite happy for the number of reported digits to depend on
> the precision which NTP reports, but to keep things simple I suggested
> using the same reporting precision as is used in the loopstats, The
> present integer microseconds are no longer adequate for the faster and
> better of today's NTP systems.
If your system had a precision of -22, I could understand your annoyance
that the reporting was just to usec. But since it is -19, I have much
less understanding of why you are getting upset. That is what I am
trying to figure out. You claim that if the report were say precision
-22 you would get more useful information. I am having trouble following
your reasoning.
More information about the questions
mailing list