IBIS Macromodel Task Group

Meeting date: 22 Apr 2008

Members (asterisk for those attending):
  Ambrish Varma, Cadence Design Systems
* Anders Ekholm, Ericsson
* Arpad Muranyi, Mentor Graphics Corp.
  Barry Katz, SiSoft
* Bob Ross, Teraspeed Consulting Group
  Brad Griffin, Cadence Design Systems
  David Banas, Xilinx
  Donald Telian, consultant
  Doug White, Cisco Systems
  Essaid Bensoudane, ST Microelectronics
  Ganesh Narayanaswamy, ST Micro
  G*** Guan, Sigrity
  Hemant Shah, Cadence Design Systems
* Ian Dodd
  Joe Abler, IBM
* John Angulo, Mentor Graphics
  John Shields, Mentor Graphics
  Ken Willis, Cadence Design Systems
  Kumar, Cadence Design Systems
  Lance Wang, Cadence Design Systems
  Luis Boluna, Cisco
* Michael Mirmak, Intel Corp.
* Mike LaBonte, Cisco Systems
  Mike Steinberger, SiSoft
* Mustansir Fanaswalla, Xilinx
  Patrick O'Halloran, Tiburon Design Automation
  Paul Fernando, NCSU
  Radek Biernacki, Agilent (EESof)
* Randy Wolff, Micron Technology
  Ray Comeau, Cadence Design Systems
  Richard Mellitz, Intel
  Richard Ward, Texas Instruments
  Sam Chitwood, Sigrity
  Sanjeev Gupta, Agilent
  Shangli Wu, Cadence
* Sid Singh, Extreme Networks
  Stephen Scearce, Cisco Systems
  Steve Pytel, Ansoft
  Syed Huq, Cisco Systems
  Syed Sadeghi, ST Micro
* Terry Jernberg, Cadence Design Systems
  Todd Westerhoff, SiSoft
  Vikas Gupta, Xilinx
  Vuk Borich, Agilent
* Walter Katz, SiSoft


--------------------------
Call for patent disclosure:

- No one declared a patent.

-----
Opens:

Has BIRD 107 been submitted?
- Walter sent this last week.
- There was one revision to fix the document title property.
- Some confusion over whether this is an official BIRD submission.
- Michael M. will post the last version received.

AR: Arpad will send to Michael M and Bob

-------------
Review of ARs:

- Walter send "final" BIRD 107 documents
  - Done

- David Banas report Xilinx position on LTI assumption for SerDes
  - No update

- Arpad:  Write parameter passing syntax proposal (BIRD draft)
  for *-AMS models in IBIS that is consistent with the
  parameter passing syntax of the AMI models
  - TBD

- TBD:    Propose a parameter passing syntax for the SPICE
  - [External ...] also?
  - TBD

- Arpad:  Review the documentation (annotation) in the macro libraries.
  - Deferred until a demand arises or we have nothing else to do

-------------
New Discussion:

Walter showed his Electrical Module Description presentation:
- Started at slide 6 "Module Interconnect Modeling Requirements"
- Is this flat or hierarchical?
  - Walter: when and EMD can call an EMD it is hierarchical
  - A backplane EMD may have board EMDs, which have DIMM EMDs, etc.
- Arpad: This discussion took place in ICM meetings
  - How does EMD fit with ICM?
  - Michael M. had proposed using ICM for module interconnect
- Walter showed a 10x30 PCB connector example:
  - Walter: ICM can not model arbitrary collections of pins
  - Arpad: can generate S-param for rectangles and stub off some pins
  - Michael M: S-params and swathing can not be used together
    - This was an oversight.
    - Could be changed quickly by eliminating one sentence.
  - Walter: why has ICM not been used for any "real" connector
  - Ian: EDA vendors have been slow to implement
  - Walter: connector companies want to release S-param models.
  - Michaal M: The proposal seems to be to replace ICM with a new superset.
  - Walter: People usually ask only for nearest neighbors and 2nd nearest.
- Arpad: We need a consensus before we render ICM irrelevant
  - Walter: We decided on EMD over ICM weeks ago
  - Michael M: The discussion was about package model support
    - Walter: We decided to leave ICM off the table until EMD was settled
  - Walter: ICM handles cables well
  - Arpad: EMD was for instantiation of lower level entities.
    - Now it seems to be becoming a universal format
  - Bob: it is not clear if EMD would include ICMs
  - Walter: ICM could describe something between EMDS, but not inside
- Arpad: This would be a bigger project because we would have to insure
  that EMD does everything ICM can.
  - Walter: EMD already does everything ICM does.
- Walter: Try creating the 10 pin S-param model in ICM
  - John: I already implemented S-param in ICM and posted it.
  - Walter: Have to create 2 node maps in ICM
  - John: This allows tools to automatically map pins to S-param ports
- Michael M: A lot has been invested in ICM
  - We were getting away from SPICE, but seem to be going back.
  - Walter showed an improvised SPICE model representation of port mapping.
- Michael M: There would be "burden shifting":
  - Burden is currently on model maker to define mapping
  - Burden would shift to system designer to do mapping
- Walter showed row/column syntax
  - Bob: this looks like swathing
  - Walter: Swathing a center model is not accurate at the edge
- Michael M: We have a need for 2 things:
  - 1 EMD
  - 2 Expanded and improved connector models
  - These should be separate efforts
- Arpad: Would prefer to use existing ICM where possible
  - Walter: EMD is for packaging, not connectors

- Michael M: We need a list of what ICM needs
  - Bob: It should have swathing of S-params
- Walter: EMD is essentially done, except:
  - Definition of S-param, etc.
  - John: EMD needs to handle coupling

Arpad and Michael M. will be traveling next week
- John Angulo will host the meeting

Next meeting: 29 April 2007 12:00pm PT

-----------