[802.3ae_Serial] From serial PMD call 26 Feb: new current issue list
Present
-------
Piers Dawe Agilent SPG
Raj Savara, John Stout Network Elements
Stretch Camnitz Agilent T&M
Tom Lindsay Stratos
Petar Pepeljugoski IBM
Greg LeCheminant Agilent T&M
Current issue list
-----------------
1. Transmitter
Transmit power setup window is very narrow in some PMDs.
Triple Trade off complexities: we now have TTO for Tx OMA and for TDP,
except 10GBASE-E. Is this really necessary?
Need to build up a track record on TDP measurements: how repeatable are
they?
For education: does TDP correlate with jitter measurements?
Should TDP measurement involve a forced decision timing window? Group
thought that it was feasible to do so by moving the decision time on the
error detector.
Is the standard receiver for TDP measurements (a) adequately designed (b)
realizable? (See Tom's Raleigh slides.)
Is eye mask test at (virtual) TP3 alone, satisfactory?
For 10GBASE-S: No. Effectively, measures Tx through a very low
bandwidth. Would need to measure at TP2 as well. What exactly is the
transversal filter supposed to do?
For 10GBASE-L: Probably, because think it won't be much different to
TP2. If so, why bother?
For 10GBASE-E: Probably not, without review and possible change to
10GBASE-E mask. Probably better to go to TP2.
Does the filter to generate the virtual TP3 for 10GBASE-S have to be a
transversal filter? Would another type be more convenient?
Bessel Thomson filters are not as accurate as we would wish.
2. Receiver
Stressed eye generation: how much stress from filter, how much from
interferer?
Stressed eye is not like the pulse splitting from poor MMF - what to do?
How do we decide how much "W" jitter the stressed eye should contain?
Exactly what are the metrics for the stressed eye?
Bearing in mind the previous week's discussion, is the 5 GHz filter right?
If not, what value?
3. Both
802.3ae D4.1 PRBSs are defined oppositely to O.150: same order of bits, but
every 1 is a 0, and vice versa.
Is the relation between TDP spec limit and amount of stress in stressed eye
in D4.1 (a) clear (b) right? Try it!
Experimentally, what is the relation between TDP and VECP of the stressed
eye?
Transmitter overshoot and high receiver bandwidth: are all Tx and Rx
interoperable? See last week's notes. [Scribe's afterthought: this
question is like a "new feature". We can't spend forever polishing the test
methods, we have a standard to publish. This question may have to be
addressed on another timeline.]
Housekeeping update of link model is desired: see below.
*** What have we missed? If we agree answers to the list above, are we
done? ***
Here's the wish list for a link model update:
Clearer handling of additional loss allowances
Other similar? requests from Tom & Paul
Support of mixed-pattern definition of extinction ratio and OMA
1 dB power change at 1550 nm
The meeting was ended abruptly by a telephone disconnection.
Next phone meeting
------------------
This will be the last teleconference before the next comment deadline and
face-to-face meeting.
The next PMD teleconference will be on Tuesday 5 March, at the usual time
and coordinates:
4:15 pm GMT = 17:15 CET = 11:15 am EST = 8:15 am PST
+1(816)650-0631 Access code 39209
Piers