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

[10GMMF] Reminder for TP2 call 4/28/05

We agreed to have a call today. I've been swamped with comments and other matters, so have nothing to present. Vivek Telang has done more work on modeling TWDP with some of John Ewen's new stressors and has a presentation. So, we should review that and discuss anything else that comes up.
 
Dial-in
  • Date: Thurs, Apr 28, 05
  • Time: 9:00 AM Pacific
  • Duration: 1:30 max
  • Number: 401-694-1515
  • Access code: 421721#
Proposed agenda
  • Attendance
  • Agenda
  • Review previous minutes (see below)
  • Proposed discussions
    • TWDP, real waveforms, etc.
      • New Vivek presentation with new stressors
      • Need more waveforms over temperature, etc.
      • Cubic spline vs. sin(x)/x interpolation and resampling (Pavel)?
    • Other/carryover
      • AC coupling (for RIN test)
      • Jitter waveform - any more work on which edges?
      • Peak power limit?
      • Test patterns
  • Next call 5/5/05
 
Thanks, Tom
 
----- Original Message -----
Sent: Monday, April 18, 2005 10:20 PM
Subject: [10GMMF] Minutes for TP2 call 4/14/05

Attendance (no order)
  • Albrecht Rommel
  • John Abbott
  • Bharath Jagannathan
  • Norm Swenson
  • Roy Blake
  • John Ewen
  • Ernie Bergmann
  • Mike Dudek
  • Piers Dawe
  • Joe Gwinn
  • Vivek Telang
  • Tom Lindsay
  • Andre Van Schyndel
  • Pavel Zivny
  • Ali Ghiasi
Agreed agenda (as per announcement)
No comments on previous minutes
Discussions
  • TWDP, real waveforms, etc.
    • Still need more waveforms that represent real Tx behaviors (over temperature, etc.). Not clear that current budgeting includes what is required.
    • Action to Tom and Vivek to run some measured waveforms over a few of the proposed TP3 stressors.
    • Premature to determine how TWDP limits should be applied to finite EQs - depends on above items.
    • Action to Pavel to compare sin(x)/x and cubic spline interpolation and resampling methods. Standard should recommend one or two. Linear interpolation is not suitable. Piers to better understand and report what Agilent did when they observed differences between 8 and 16 samples per UI. ClariPhy studied this with cubic spline and found little sensitivity down to less than 4 samples per bit.
  • Other/carryover
    • Back reflection - agreement by Mike and Piers that test should use MMF and multi-mode limits (-12 dB). They will comment during current ballot.
    • AC coupling (for RIN test) - Tom mentioned backplane work that showed 4.7 nF as optimum (330 kHz), but not conclusive because it used on prbs7 data.
    • Jitter waveform - Pavel reported observations that all edges are very close. He will try to report results on a Tx they have.
    • Test patterns - not discussed.
    • Noise bandwidth and k factor - not quite sure what to report here. Noise measurements should be integrated over the actual frequency response of the measurement system; if noise is white, then noise bandwidth is valid, which is ~5% higher than the 3 dB bandwidth for a Bessel Thomson filter. k factor is used in MPN prediction, and the 10GE spreadsheet also used a RIN factor. Suggestion to review current draft for problems in these areas.
Next call - 2 weeks.
  • Date: Thurs, Apr 28, 05
  • Time: 9:00 AM Pacific
  • Duration: 1:30 max
  • Number: 401-694-1515
  • Access code: 421721#
 
Please send comments or corrections.
Thanks,
Tom Lindsay
ClariPhy Communications
tom.lindsay@clariphy.com
phone: (425) 608-0209 or (949) 480-9210
cell: (206) 790-3240
fax: (425) 608-0232
 
----- Original Message -----
Sent: Wednesday, April 13, 2005 3:21 PM
Subject: [10GMMF] Reminder for TP2 call 4/14/05

TP2'rs -
 
Details for Thursday's call
  • Date: Thurs, Apr 14, 05
  • Time: 9:00 AM Pacific
  • Duration: 1:30 max
  • Number: 401-694-1515
  • Access code: 421721#
Proposed agenda
  • Attendance
  • Agenda
  • Review previous minutes (see below)
  • Proposed discussions
    • TWDP, real waveforms, etc.
      • Need more waveforms over temperature, etc.
      • TP3 has stressor candidates - any analysis with real waveforms?
      • Penalty limits for finite EQs?
      • Cubic spline vs. sin(x)/x interpolation and resampling (Pavel)?
    • Other/carryover
      • Back reflection
      • AC coupling (for RIN test)
      • Jitter waveform - any more work on which edges?
      • Test patterns
  • Next call
 
Tom