Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index


There have been a number of questions of late about jitter and the
interpretation of the specifications.

For 8B/10B (XAUI; WWDM...)

 It should be recognized that there is good probability that we will use the
same jitter methodology used in 802.3z for the XAUI interface and the WWDM
PMD. Much of the state-of-the-art in 8B/10B jitter measurement and
specification was included in 802.3z. Additional information can be found on
the Fibre Channel site in the following documents:

I would highly recommend these documents. If discussion on the reflector
seems light on this topic, it might well be because there is about 5 years
of history here (work kicked off at June, 95 FC meeting in Rochester, MN)
that isn't very easy to communicate on a reflector. I sure wouldn't want to
attempt it.

For 64b/66b (Serial PMDs)

Here we are going to be on some new ground. Hopefully, we will be based on
the same basic principles and theory found in the work mentioned above plus
some of the early jitter work done in SONET. There are a number of us that
will be getting together at the meeting sponsored by the 10GEA to discuss
this next week (see the meeting minutes, page 1 from Sept, 2000 on the web).

I would expect that the majority of discussion related to this would occur
on the new SerialPMD reflector.


Jonathan Thatcher, Principal Engineer
Chair, IEEE P802.3ae (10 Gigabit Ethernet)
World Wide Packets
115 North Sullivan Rd
PO Box 950
Veradale WA  99037
509-242-9228; Fax 509-242-9001;