IBIS Macromodel Task Group

Meeting date: 19 October 2010

Members (asterisk for those attending):
Agilent:                    * Fangyi Rao
                              Radek Biernacki
Ansoft:                       Chris Herrick
                            * Danil Kirsanov
Cadence Design Systems:       Terry Jernberg
                            * Ambrish Varma
Celsionix:                    Kellee Crisafulli
Cisco Systems:              * Mike LaBonte
                              Stephen Scearce
Ericsson:                   * Anders Ekholm
Intel:                      * Michael Mirmak
LSI Logic:                    Wenyi Jin
Mentor Graphics:            * John Angulo
                              Vladimir Dmitriev-Zdorov
                              Zhen Mu
                            * Arpad Muranyi
Micron Technology:            Randy Wolff
Nokia-Siemens Networks:     * Eckhard Lenski
Sigrity:                      Brad Brim
                              Kumar Keshavan
                            * Ken Willis
SiSoft:                     * Walter Katz
                              Mike Steinberger
                            * Todd Westerhoff
ST Micro:                     Syed Sadeghi
Teraspeed Consulting Group: * Scott McMorrow
                            * Bob Ross
Vitesse Semiconductor:        Eric Sweetman
Xilinx:                       Mustansir Fanaswalla

(NOTE: The list above is been pruned to include only those attending at
       least one meeting in the past year, and reorganized by company.)

The meeting was lead by Arpad Muranyi

------------------------------------------------------------------------
Opens:

- Arpad: Created an agenda from the task list
  - We may not get to all items
  - We should discuss BIRD 119
  - Ken will have to drop off early
- Walter: I can give an overview of BIRD 119 in 5 minutes

- Mike L: Would like to address task group protocol changes suggested by
  Michael Mirmak
  - Mike sent an attendance count list to Arpad
  - The attendance list should be pruned to exclude attendees beyond 1 year
  - Also organize by company

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

- none

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

- Arpad rework Typos BIRD draft to include Walter's ideas about documentation
  of Format
  - Done

- Arpad address BIRD 114.2 escape sequences in strings
  - This will be discussed offline

- 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
  - Done, BIRD 117 & 118 submitted

- TBD:    Propose a parameter passing syntax for the SPICE
          - [External ...] also?
  - Done, BIRD 117 & 118 submitted

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

Typos BIRD discussion:
- Ambrish: Have we gone over the other BIRDs?
- Arpad: They have been submitted
  - We still can discuss

Walter showed the BIRD 119 Reserved Parameters draft:
- This is based on 2 years of problem solving
- Many IC vendors have tools based on Matlab, etc.
- They have to bring in other files for these tools
- The BIRD proposes Supporting_Files, DLLPath, DLLid, Samples_Per_Bit
- Also parameters for Broadband Analog, Equivalent-Circuit, Jitter
- Added parameter dependency tables
- This BIRD is related to Arpad's BIRDs 116-118

- Mike: The Equivalent-Circuit Analog parameters are a migration of simple
  data types from the IBIS file into the AMI file
- Walter: True
- Fangyi: Does this mean we can bypass IBIS files?
- Walter: IBIS handles things AMI doesn't yet, like diff pins
  - Several other companies are also requesting these changes
- Todd: Migrating these parameters is a pragmatic step
- John: The IBIS buffer model only has [Model Selector] for variations
  - That can get very verbose

- Arpad: This assumes there is a template circuit somewhere
  - IBIS-ISS would be less hard-coded
  - The AMI file could still parameterize it

- Arpad: Should this be broken into smaller BIRDs?
- Walter: A number of existing models need all of this
  - The ideas came from Opal
  - It would be best to vote on them as a block
- Scott: The entire BIRD?
  - Teraspeed would object
  - This BIRD crosses multiple domains
- John: We should not let "political" issues dictate our pace
- Walter: It could be 4 BIRDs
- Scott: I don't want to end up with more spec changes in the future
- Walter: We can't foresee how Jitter will be handled for 10 years, for
  example
  - We did not understand how IC vendors would model jitter in their tools
  - EDA tools should reflect how they model
- Scott: Mostly concerned about equiv analog modeling
  - The boundaries can be mushy
  - IC vendors can have dumb models that were mostly easy to do in Matlab
- Walter: The Si designer has created analog partitions we can't change
  - Their algorithmic and analog models have to play together
- Scott: The spec should be clear about what goes into each side
- Walter: Reactive loads make that hard
- Fangyi: It should define how analog will work with the IBIS model
- Walter: I can either withdraw BIRD 119 or make it one of the 4

AR: Walter divide BIRD 119 into 4 separate BIRDs

Arpad showed the Typos BIRD draft:
- Changed format to be more formal
- Changed Format to use symbols in <> instead of spelling out all options
- Example "Format Table" is now just "Table"
- Fixed indentation
- Made the implicit default value for List to be the Typ value
- Arpad: It could simply say the Typ is the implicit default for all types
- Arpad: Should Description be required or optional for Model Specific?
- Todd: It should be required
- Walter: It should be optional with a recommendation to include it
- Bob: Agree with Walter
- Arpad: I made a change to say Model_Specific is not required
- Arpad: Had to add three examples to show interchangeable Default and Value
  - The third example has Format Value
- Walter: We have described parameter requirements in one place
  - We could say they're not List and leave it at that
- Bob: Currently Description is required
- Arpad: Added two examples for Tx_Jitter and Tx_DCD
  - One has Format added
  - Someone had asked for this
  - Other examples have Format removed
- Walter: Tx_Jitter as Table is useful
  - Other formats are not useful for that
  - This will be in my BIRD
- Arpad: Format has been removed from other examples
  - Similar changes made to other Rx parameters
- Walter: We should clarify column types in Table examples
- Arpad: The first column is always a name value
- Bob: We need to revisit the whole issue of Table
  - There is no restriction, it can be used for anything
- Arpad: Need to consistently spell "Data Format"
- Arpad: The parser developer says there is a conflict between text and format
  - Not sure about the details

AR: Arpad change Typos BIRD to say the Typ is the implicit default for all types
AR: Arpad and Walter discuss parameter Default/Value examples

-------------
Next meeting: 26 October 2010 12:00pm PT

Next agenda:
- BIRD 119
- Any other new BIRDs

-------------
IBIS Interconnect SPICE Wish List:

1) Simulator directives

