[time-nuts] Basic Stratum 1 question

[Warner Losh]

From: "Jared Morrisen" <jaredmorrisen at gmail.com>
Subject: [time-nuts] Basic Stratum 1 question
Date: Thu, 2 Aug 2007 11:32:06 -0400

> ); SAEximRunCond expanded to false
> Errors-To: time-nuts-bounces+imp=bsdimp.com at febo.com RETRY
> 
> Hi,
> 
> I am having a debate with our CIO.  He wrote in a memo about timing:
> 
> *Local hardware is to be considered Stratum 1, since it get time from its
> own CMOS.*
> 
> 
> 
> I told him that absurd and that it can't be considered stratum 1.
> 
> 
> 
> Please clarify.

Stratum 1, in the ntp world, is a source that's connected to an
official source of time (the so-called Stratum 0).  The CMOS clock
isn't connected to NIST, therefore it can't be considered Stratum 1 by
any stretch of the imagination.

Also, Stratum 1, in telco, has certain performance requirements.
These timing requirements are relatively stringent.  Even using an
insanely generous specification of 1ms and 1ppm, an unsciplined PC
clock cannot meet that spec for more than a few seconds given that the
typical frequency error is on the order of ~100ppm.

In short, your CIO is mistaken[*].

Warner

[*] Never attribute to malice what can adequately be explained
by incompetence.