[ntpwg] Testing NTP performance

[Stewart Bryant]

Stuart
>
> Asked another way, folks here have claimed that NTP runs
>  with particular accuracies much tighter
>    than a human can check by printing out the time on a console.
>
>   
My guess is that they make a hardware measurement, it is certainly the 
most accurate way to do it.

In the 1588 bakeoffs I have seen the most popular technique was to look 
at the eye pattern, but I am fairly sure that there is commercial 
equipment that takes the pulses and  measures  mtie, tdev etc, although 
there is concern as to whether these are the most appropriate metric in 
a packet as oppsoed to a synchronous environment.

- Stewart