Re: [802.3_100GCU] Suggested practices for reporting simulation results

["Mellitz, Richard" <richard.mellitz@xxxxxxxxx>]

Are we assuming all Tx jitter is high frequency?

… Rich

 

From: ZIAD HATAB [mailto:zrhatab@xxxxxxx]
Sent: Friday, February 18, 2011 3:07 PM
To: STDS-802-3-100GCU@xxxxxxxxxxxxxxxxx
Subject: [802.3_100GCU] Suggested practices for reporting simulation results

 

Adam,
 
Here's my feedback:
 
1) Parameters to be included in addition to your list:

Link: Bit Encoding (e.g. 64B/66B. This is probably the default value)
Transmitter: pre-emphasis technique and levels (e.g. 2-tap FIR or 3-tap FIR)
Receiver: Equalization technique and number of taps (e.g. LMS with fractionally-spaced FFE)
Channel: Number of NEXT and FEXT sources. Package Crosstalk.

 
2) Parameters needing clarification:

Tx and Rx deterministic jitter distribution: are you referring to normal or uniform etc.? If so, is it necessary?

 
Also, I think that having the channel single-ended resistance and capacitance values is a good idea.
 
Please add me to the upcoming teleconference on this topic. Thanks,
 
Ziad Hatab
Vitesse
 

 

From: Healey, Adam B (Adam) [mailto:Adam.Healey@xxxxxxx]
Sent: Thursday, February 17, 2011 5:47 PM
To: STDS-802-3-100GCU@xxxxxxxxxxxxxxxxx
Subject: [802.3_100GCU] Suggested practices for reporting simulation results

 

Colleagues,

 

At the January 2011 interim meeting of the IEEE 802.3 100 Gb/s Backplane and Copper Cable study group, I accepted an action item to…

 

“Lead discussion (reflector or ad hoc) to establish practices for reporting simulation results by March meeting.”

<http://www.ieee802.org/3/100GCU/public/jan11/minutes_01_0111_unapproved.pdf>

 

This was assigned in the context of the presentation “Suggested practices for reporting simulation results” which proposed that we foster consistent and complete disclosure of the simulation conditions.

<http://www.ieee802.org/3/100GCU/public/jan11/healey_01_0111.pdf>

 

The aim is to summarize the essential parameters that influence the result of a simulation and request that the values of those parameters be reported so that simulation results may be understood in their full context. A starting point for such a list was included in the aforementioned presentation.

 

This e-mail is being sent to solicit feedback on that list. I would expect feedback to take any of the following forms.

a) Parameters whose definition needs to be clarified.

b) Parameters to be included

c) Parameters to be omitted

 

I would like to stress that is not an objective to set specific values for any of these parameters at this time, or dictate what must be included in a simulation for it to be considered “valid”. This is simply a list of recommended items to be included in a report, e.g. whether or not they are included and what value was used for those that are included.

 

In addition to your feedback, I would like to earmark time for a teleconference to consolidate thoughts on this matter and prepare an update for the study group meeting scheduled to take place during the March plenary week. I propose that this meeting take place Thursday, March 3 at noon EST (9 am PST). If you would be interested in participating in this meeting, please let me know so that I can allocate teleconference bridge ports accordingly.

 

In the meantime, I look forward to any feedback you might have on this topic. Please utilize the reflector for discussion.

 

Thank you,

 

Adam Healey

LSI Corporation