Re: [BP] Parameter input to signaling spreadsheet

[Brian Brunn <Brian.Brunn@xxxxxxxxxx>]

Hi Joe,

 

My thought for including the tap values was to offer some ability for independent verification of the results including architecture complexity.  Not to acquire proprietary information.

 

With the board/connector vendors, we have asked that they make s-parameter data available while we have not required them to specifically show how they achieved these results.  Similarly, I was thinking that supplying the tap values for a few channels does not explicitly reveal the algorithm used to generate these tap values.

 

Would an acceptable alternative be to supply an equalized pulse response?

 

Thanks,

Brian

 

 

----- Original Message -----

From: Joe M Abler

To: STDS-802-3-BLADE@listserv.ieee.org

Sent: Tuesday, November 30, 2004 1:29 PM

Subject: Re: [BP] Parameter input to signaling spreadsheet

I do not want to include tap value settings.  We consider our algorithm for optimizing these values to be proprietary, and perhaps other vendors will as well.  I'm sure it would be interesting additional information were everyone to provide it, but I don't believe it would weigh on a decision process.  

Thanks,        Joe


Joe Abler                                                             abler@us.ibm.com
IBM Microelectronics Division                          919-254-0573
Technical Marketing & HSS Applications    919-254-9616 (fax)
3039 Cornwallis Road                                                                
Research Triangle Park, NC  27709

Brian Brunn <Brian.Brunn@XILINX.COM> Sent by: owner-stds-802-3-blade@listserv.ieee.org 11/30/2004 10:49 AM Please respond to Brian Brunn

To:        STDS-802-3-BLADE@listserv.ieee.org         cc:                 Subject:        [BP] Parameter input to signaling spreadsheet

Additional parameters for signaling spreadsheet

(1) Input referred device noise

(2) Input referred offset

(3) Input referred environmental noise

(4) Reference level sampling error

(5) Reference level generation error

(6) Ensuring that eye-width is measured at the 1/2 eye height point.

(7) Listing of all tap values after adaptation

(4) and (5) are specific imperfections that result from multi-level

signaling.  To set a multi-level threshold, a circuit needs to measure

something (4) and a circuit needs to use the measurement to generate

the threshold (5).  Both of these will have errors.

(6) is also a multi-level issue.  Multi-level eyes are symmetric in

voltage and time but not simultaneously.  The widest eye-width is not

at the center of the eye-height.  We need to ensure the results are

not the result of taking the best of both.

Thanks,

Brian T. Brunn

Xilinx, Communications Technology Division

805 Las Cimas Pkwy #250

Austin, TX  78746

brian.brunn@xilinx.com

512-634-1304