[ntp:questions] precision in a driver

Unruh unruh-spam at physics.ubc.ca
Fri Jan 4 21:06:29 UTC 2008


"Richard B. Gilbert" <rgilbert88 at comcast.net> writes:

>Folkert van Heusden wrote:
>> Hi,
>> 
>> The precision-field set by a driver, what should one set there? If a
>> clock gives ms values, -10? Or is it the real quality of the
>> time-source? Like: maybe that on average the source has a quality of
>> 10ms.
>> 
>> 
>> Folkert van Heusden
>> 

>The precision is the smallest possible difference between two 
>consecutive requests for the time!  It's specified as a negative power 
>of two.

On most systems now adays it is of the order of microseconds (-19 or -20)







More information about the questions mailing list