Minutes, IBIS Quality Committee

23 Sep 2008

11-12 AM EST (8-9 AM PST)

ROLL CALL
  Adam Tambone
* Anders Ekholm, Ericsson
  Barry Katz, SiSoft
  Benny Lazer
  Benjamin P Silva
  Bob Cox, Micron
* Bob Ross, Teraspeed Consulting Group
  Brian Arsenault
* David Banas, Xilinx
* Eckhard Lenski, Nokia Siemens Networks
  Eric Brock
  Guan Tao, Huawei Technologies
  Gregory R Edlund
  Hazem Hegazy
  Huang Chunxing, Huawei Technologies
  John Figueroa
  John Angulo, Mentor Graphics
  Katja Koller, Nokia Siemens Networks
  Kevin Fisher
  Kim Helliwell, LSI Logic
  Lance Wang, IOMethodology
  Lynne Green
* Mike LaBonte, Cisco Systems
  Mike Mayer, SiSoft
* Moshiul Haque, Micron Technology
* Pavani Jella, TI
  Peter LaFlamme
  Randy Wolff, Micron Technology
  Radovan Vuletic, Qimonda
  Robert Haller, Enterasys
  Roy Leventhal, Leventhal Design & Communications
  Sherif Hammad, Mentor Graphics
  Todd Westerhoff, SiSoft
  Tom Dagostino, Teraspeed Consulting Group
  Kazuyoshi Shoji, Hitachi
  Sadahiro Nonoyama

Everyone in attendance marked by *

NOTE: "AR" = Action Required.

-----------------------MINUTES ---------------------------
Mike LaBonte conducted the meeting.

Call for patent disclosure:

- No one declared a patent.

AR Review:

- Moshiul rewrite 5.4.5 to describe leading waveform delay requirements
  - Done, discussed today.

New items:

Continued review of the IBIS Quality Specification:

- Mike: The currently posted IQ version is the 1.1ah draft
  - Mike has a 1.1ai version that will be posted after today's edits.

5.4.5.	{LEVEL 3}  Output and IO buffers should have 2 V-T tables
- Checks 5.4.2 and 5.4.5 combined into one
- The language has been clarified not to demand a 50 ohm R_fixture
- IBISCHK checks final value settling value between waveforms and [Ramp]:
  - 2% offset generates a warning
  - 10% offset generates an error
- The same amount of leading delay must be removed for all min curves:
  - This applies to typ and max, but independently.
- Is 1 waveform OK?
  - The simulator has to guess the K-factor curve in this case
    - Variation between simulators is likely
    - Bob: HSPICE has a dynamic algorithm for the K-factor curve
  - Mike: One waveform is allowable where the termination scheme is pre-defined
  - Bob: Terminated topologies still can have 2 waveforms
    - The fixture voltages should be just above and below center
  - Moshiul: Only 1 waveform is needed for open-source and open-drain
- Bob: I/V curves sometimes are far away from buffer switching characteristics
  - They represent end-point conditions only
- Bob: We should not specify too much in the IQ checks
  - Compliant models may be deemed bad
- Anders: Sometimes it is recommended to have only [Ramp], no waveforms
  - This is mostly for very high speed linear devices
- Anders: Ramp should be checked against V-T data
  - In many models the [Ramp] data can be derived from waveforms
  - Not all models use the same fixture R and V as for [Ramp], but some do
- Is this really a level 3 check?
  - It would be a level 2 check if problems would impact waveform simulation
  - It is level 3 if problems only impact timing
  - Mike: With V-T tables commented out, the simulation results change

Next meeting:

30 Sep 2008 11-12 AM EST (8-9 AM PST)

Meeting ended at 12:01 PM Eastern Time.