[10GMMF] Reminder for TP2 call 2/10/05

[Tom Lindsay <tom.lindsay@xxxxxxxxxxxx>]

Next call will continue discussions started last week. See items below in bold.

Date: Thurs, Feb 10, 05 (regular day/time)

Time: 9:00 AM Pacific

Duration: 1:30 max

Number: 401-694-1515

Access code: 421721#

Tom

 

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

From: Tom Lindsay

To: STDS-802-3-10GMMF@listserv.ieee.org

Sent: Saturday, February 05, 2005 11:34 AM

Subject: [10GMMF] Minutes for TP2 call 2/3/05

Attendees (no order)

• John Ewen
• Vivek Telang
• John Jaeger
• Mike Dudek
• Piers Dawe
• Petar Pepeljugoski
• Tom Lindsay
• Lars Thon
• John Abbott

Agenda

• Attendance
• Agenda
• Review old minutes (1/20)
• TP2 topics (see below)
• Next meeting/call

TP2 topics

• TWDP test pattern (from AnAi or BnBi)
  • Important effort.
  • John Ewen still interested in helping.
• Scope noise and jitter compensation method/sentences
  • Use simple rss subtraction of O/E & scope contribution.
  • Tom - add as comments into next draft.
• Clock recovery/tracking
  • Uncorrelated jitter test
    • Should track per 4 MHz high pass spec when receiving normal transition density traffic.
    • Square wave not friendly to CDRs, document should at least allow use of other test patterns with jitter measured on selected edge.
    • Tom - pass this issue back to instrumentation folks.
  • TWDP test
    • AnAi and BnBi repeat ~every 16986 bits, but the capture should begin at the same point and include the same bits. That is, the overall pattern length is not the issue, but whether there are long-term jitter variations within the capture range. The capture range is expected to be ~500 bits.
    • prbs9 is 511 or 512 bits.
    • Time constant of a 4 MHz tracking filter is ~400 bits, so test patterns are more than 1 time constant. They are long enough that perhaps there is some need for tracking in the expected presence of jitter.
    • Watch this and see if a need comes up from analysis or testing.
    • The presence of RJ will slow edges during averaging. This will lead to a higher penalty calculation, which is appropriate.
• RIN
  • Should the Rx be running during the test? (Such words are in the new jitter test). Answer is essentially yes: Add words to the effect that it should be running if it makes a difference.
  • Allow AC coupling during RIN? (Reduce noises that would be normally filtered by AC coupling). RIN, per 802.3ae, includes AC coupling below 1 MHz. This is adequate, be sure LRM includes the same.
• MATLAB code review
  • John and Vivek attempting independent calculations. John within +/-0.2 dB. Tom to send linear 47.1 psec Gaussian waveform (used in ClariPhy work to compare to PIE-D as shown in Vancouver) to John for comparison. Done 2/4.
  • Tom to send prbs9 patterns to Piers (both 511 and 512 bit versions). Done 2/4.
  • How many channels/fibers are required for TWDP test? May be less than for TP3, since it uses an ideal (very long) equalizer. Not concluded, but goal is to minimize.
  • NEW - discussion about threshold - mid vs. mean?
• More TP2 waveforms - do we have the right TWDP and TP3 values? Other response shapes?
  • Suppliers need to evaluate transmitters, receivers, waveshapes, etc. to be sure we draw the line correctly. TWDP and TP3 limits currently ~5 dB.
• Rx overload (driven by Rx folks?)
  • Watch and respond as required.

Next call

Date: Thurs, Feb 10, 05 (regular day/time)

Time: 9:00 AM Pacific

Duration: 1:30 max

Number: 401-694-1515

Access code: 421721#

 

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 -----

From: Tom Lindsay

To: STDS-802-3-10GMMF@LISTSERV.IEEE.ORG

Sent: Wednesday, February 02, 2005 9:42 PM

Subject: [10GMMF] Reminder for TP2 call 2/3/05

Sorry about the late reminder.

• Date: Thurs, Feb 3,05 (regular day/time)
• Time: 9:00 AM Pacific
• Duration: 1:30 max
• Number: 401-694-1515
• Access code: 421721#

For the call, I'd like to brainstorm a list of TP2 topics (a starting list is below) that require work, and determine how/when we want to tackle them.

• TWDP test pattern (from AnAi or BnBi)
• Scope noise and jitter compensation method/sentences
• Clock recovery/tracking
  • During uncorrelated jitter test?
  • During TWDP test?
• RIN
  • Should the Rx be running during the test? (Such words are in the new jitter test.)
  • Allow AC coupling during RIN? (Reduce noises that would be normally filtered by AC coupling.)
• MATLAB code review
• More TP2 waveforms - do we have the right TWDP and TP3 values? Other response shapes?
• Rx overload (driven by TP3 folks?)
• Others?