IBIS Macromodel Task Group

Meeting date: 03 August 2010

Members (asterisk for those attending):
  Adge Hawes, IBM
* Ambrish Varma, Cadence Design Systems
  Anders Ekholm, Ericsson
* Arpad Muranyi, Mentor Graphics Corp.
  Barry Katz, SiSoft
* Bob Ross, Teraspeed Consulting Group
  Brad Brim, Sigrity
  Brad Griffin, Cadence Design Systems
  Chris Herrick, Ansoft
  Chris McGrath, Synopsys
* Danil Kirsanov, Ansoft
  David Banas, Xilinx
  Deepak Ramaswany, Ansoft
  Donald Telian, consultant
  Doug White, Cisco Systems
* Eckhard Lenski, Nokia-Siemens Networks
  Eckhard Miersch, Sigrity
  Essaid Bensoudane, ST Microelectronics
  Fangyi Rao, Agilent
  Ganesh Narayanaswamy, ST Micro
  Gang Kang, Sigrity
  Hemant Shah, Cadence Design Systems
  Ian Dodd, consultant
  Jerry Chuang, Xilinx
  Joe Abler, IBM
  John Angulo, Mentor Graphics
  John Shields, Mentor Graphics
* Ken Willis, Sigrity
  Kellee Crisafulli, Celsionix
* Kumar Keshavan, Sigrity
  Lance Wang, Cadence Design Systems
  Luis Boluna, Cisco Systems
* Michael Mirmak, Intel Corp.
* Mike LaBonte, Cisco Systems
  Mike Steinberger, SiSoft
  Mustansir Fanaswalla, Xilinx
  Patrick O'Halloran, Tiburon Design Automation
  Paul Fernando, NCSU
  Pavani Jella, TI
* Radek Biernacki, Agilent (EESof)
* Randy Wolff, Micron Technology
  Ray Komow, Cadence Design Systems
  Richard Mellitz, Intel
  Richard Ward, Texas Instruments
  Samuel Mertens, Ansoft
  Sam Chitwood, Sigrity
  Sanjeev Gupta, Agilent
  Scott McMorrow, Teraspeed Consulting Group
  Shangli Wu, Cadence Design Systems
  Sid Singh, Extreme Networks
  Stephen Scearce, Cisco Systems
  Steve Kaufer, Mentor Graphics
  Steve Pytel, Ansoft
  Syed Huq, Cisco Systems
  Syed Sadeghi, ST Micro
  Ted Mido, Synopsys
  Terry Jernberg, Cadence Design Systems
  Todd Westerhoff, SiSoft
  Vladimir Dmitriev-Zdorov, Mentor Graphics
  Vikas Gupta, Xilinx
  Vuk Borich, Agilent
* Walter Katz, SiSoft
  Wenyi Jin, LSI Logic
  Zhen Mu, Mentor Graphics

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

- none

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

- none

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

- Walter: Send updated Init_Returns_Filter BIRD to list and Mike for posting
  - Done

- 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:    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:

IBIS-ISS update:
- Please provide comments, especially on structure and readability
- Arpad: Why was it posted in a new place?
- Bob: This is an official controlled location
- Bob: Did we have op-amp as a controlled source?
- Michael M: There were cross references to it
  - Both marked up and plain pages are posted to make the changes visible
- Walter: We only added DC sources
- Radek: It was to create a short circuit
- Michael M: EFGH elements all use common syntax
  - That chart was not originally included
  - The whole chart was added in
- Arpad: The idea was to support common interconnect models, so op amp is not
  needed
- Michael M: I will delete it, but other questions may be raised
- Michael M will present next week

AR: Mike L add IBIS-ISS link to work archive
AR: Michael M delete op amp from IBIS-ISS document

Ken Willis gave a presentation "Proposal for AMI Flow cases 4 & 7"
- Slide 6:
  - It is only an issue for TD simulation
  - Is it worth the added complexity?
- Slide 9:
  - With dual models users can pick which part they want to use
- Slide 10:
  - Init-based models typically allow user settings
  - The EDA tool can sweep Rx settings
  - There is a clean path for TD simulations of case 4
- Slide 11:
  - Case 4 is rare and EDA tools can deal with it anyway
  - There is no need to burden the spec to cover case 4
- Walter: Can not disagree
  - Close to my option #1 "Do nothing"
- Ken: It could also be option #5 "No explicit TD flow defined"
- Walter: The deconvolution issue is with case 7, not 4
  - Ken's proposal is acceptable
- Ambrish: If Rx Init has optimization the tool can take care of it
- Arpad: Then we will not submit the second flow BIRD?
- Ken: Agree
- Radek: We should describe what tools need to do
- Arpad: We should still give a reference flow
  - EDA tools need to come up with the same results
- Walter showed his list of deconvolution options
  - That is my #1 proposal
  - We should rank them to see what is preferred
- Bob: We need to decide on Init_Returns_Filter
- Arpad: Ken's proposal rules out options 2-4
  - Option 6 is also ruled out
  - Option 1 would win over #5
- Ambrish: Option #1 is OK with a reference flow
- Ken: Agree
- Kumar: Is #5 based on optimization?
- Ken: #1 defines a reference flow, #5 does not
- Arpad: The issue is case #7 deconvolution
- Walter: There is no way for the EDA tool to know if Init will optimize
  - A human can figure it out, but a tool can't
- Kumar: The EDA tool has to do the job either way
  - We can't prescribe one method
  - Some users want the EDA tool to do the optimization
- Ambrish: The tool does not always need to deconvolve with the default case
- Bob: Would a new Tool_Does_Optimization parameter solve this?
- Ambrish: The presence of Rx GetWave can help
- Kumar: Users should choose
  - A boolean should not be needed
- Michael M:
  - Three options:
    - No adaptive EQ (RX-TX communication)
    - Override with user values for testing
    - Compare what the silicon does to some what-if
  - How to know if the models will support these?
- Walter: Flow #1 works for all cases
- Ken: For advanced models users will have to read documentation
- Arpad: It sounds like we are settled on option #1
- Walter: We need to read it carefully
  - We might want to propose words for any problematic cases
- Arpad: At the open forum it was asked when this would be submitted
  - We could submit as is and review there
- Ambrish: I am preparing a proposal
- Bob: Could go either way
  - There are some editorial issues
- Arpad: It already states that deconvolution may be necessary
- Ken: We should alert model makers and users about the issues soon
- Arpad: The BIRD has no major technical issues, only editorial
- Bob: I have editorial comments
- Walter: Agree that we should submit
- Bob: It takes 6 weeks to become a candidate for voting
- We decided to wait 1 week before submitting

AR: Walter send a DOC format file for his flow BIRD

Next meeting: 10 August 2010 12:00pm PT

--------

IBIS Interconnect SPICE Wish List:

1) Simulator directives
