[ntpwg] ntpwg Updated NTPv4 Protocol Specification/timestamp location

David L. Mills mills at udel.edu
Thu Feb 21 04:31:16 UTC 2008


Guys,

May I offer a summary, analysis and proposal about all that has gone 
down here? Stuart Venters and I cooked it up.

The following document discusses software, hardware and driver 
timestamping schemes. It analyzes the errors with each scheme and 
combinations of schemes. It makes a specific recommendation on 
interoperation between all three schemes to achieve accuracies 
approaching IEEE 1588.

http://www.eecis.udel.edu/~mills/timestamp.txt

Dave

Brian Utterback wrote:

>
> Al Morton wrote:
>
>> Kurt,
>>
>> B set T2 at the end of the packet, so it includes serialization time.
>> B set T3 at the end of the packet, where it excludes serialization time.
>> (A sets T1 and T4 at the beginning of the packets).
>>
>> Consistency among senders and receivers seems to be important here.
>>
>> movin' on,
>> Al
>>
>
> Which was exactly the point that he was trying to show. Consistency
> is important, or at least relevant. The maximum error introduced is
> 1 frame time. But marking start of packet is much easier than end of
> packet, which involves making predictions.
>



More information about the ntpwg mailing list