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

[802.3_10SPE] Wednesday's Ad hoc - chair's remarks



Because we have a VERY FULL ad hoc agenda tomorrow, I am making my remarks by email, and I’ll keep them short.

We intend to start at 7AM (pacific) SHARP! – no “5 minute grace” is planned – please be on time.  We have the pleasure of many presenters advancing our work.

 

REMINDER – Please let me know if you plan to make a presentation at the September interim – Sooner is better than later, but NO LATER THAN Monday morning.  Remember, Monday is a holiday in the US, so do it before the weekend to be sure!

 

Please let me know even if you aren’t sure.  Presentations themselves are not due until Thursday September 7.

 

We are making progress on multiple fronts.

  1. Long Reach PHY -  Mr. Graber and his colleagues have been working on the detail of the long-reach PHY specification.  Steffen has a presentation for the ad hoc which covers some of the basics, but should have text for us to review and adopt at the interim.  He’s done a lot of work on the detail of the PCS and PMA. Please be prepared to spend time with the text so that we can modify and/or adopt it.
  2. Multidrop – we need to continue to make positive progress and make some steps forward in Charlotte.  We have 3 viable approaches I’ve seen thus far, none of which seem quite there yet.  Please provide the presenters constructive feedback.  Thus far the reflector has been quiet.  Now that many of us are back from vacation, I hope that will change.
    1. Piergiorgio has presented concepts for multidrop and gotten feedback.  We have a refinement for today which removes some of the more problematic parts of the older presentations.  We have briefly discussed some issues of project scope as to how these proposals fit with 802.3 and an 802.3 PHY project.  I think we are getting closer – let’s continue to discuss any needed modifications.
    2.  Craig has more thoughts on using 802.1 to improve scheduling performance on CSMA/CD.  Others have concern about the complexity of the overall system in cost-sensitive automotive environments.  We need to discuss those trade offs in order to understand and resolve exactly what 802.1 functionality needs to be layered and what additional burden it brings.
    3. Geoff Thompson has mentioned to me 802.3w, which was an improvement to 802.3 MAC functionality abandoned at the time switched Ethernet rises.  If someone wants to research this topic and consider it – then please go ahead.  Such an effort would almost surely be outside our current scope and would need a CFI, but might be a fruitful path forward as well.
  3. Automotive & short reach link segment specifciations –  we need these for both multidrop and point-to-point applications, we have a presentation on the automotive channel, and I believe more for Charlotte.
  4. Other use cases and short-reach PHY –
    1. We need someone to champion a complexity analysis for point-to-point short reach.  It is important to know how the complexity of an implementation for 15m stacks up against a 15m version of the currently proposed long-reach PHY, as well as a 15m 100BASE-T1 PHY.
    2. We also need detail on potential phy requirements for the use cases we’ve adopted from the NEA activity.

 

If you would like to discuss needs of the task force on any of these issues, or have ideas you might help bring 802.3cg forward, please do not hesitate to contact me privately.  Our conversations will remain between us as long as you want.  My job as chair is to progress this standard, and I am at your service.

 

-george

 

George Zimmerman, Ph.D.

President & Principal

CME Consulting, Inc.

Experts in Advanced PHYsical Communications

george@xxxxxxxxxxxxxxxxxxxx

310-920-3860