IBIS Macromodel Task Group Meeting date: 13 October 2009 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 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, Cadence Design Systems 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 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 Zhen Mu, Mentor Graphics ------------------------------------------------------------------------ Opens: - Arpad announced that Mike L. was not available today, and asked for any volunteers to take the meeting minutes. - Arpad gave an overview of the schedule for the upcoming meetings: - Vladimir's presentation on including non-linear effects (today) - Discuss Walter's E-mail "An AMI Overview", sent Sat 10/10/2009 - Discuss "Three new flow diagrams have been posted", posted Fri 10/9/2009 - Walter Katz will prepare an AMI BIRD draft - Refine AMI BIRD draft - Add Computation DLL Entry - Add LTI Reserved Parameters to AMI BIRD - Add Vendor Requested Reserved Parameters to AMI BIRD - Request for other changes to AMI BIRD - Submission of AMI BIRD to IBIS Open Forum - Make a list of IBIS-ISS questions to be answered - Discuss/resolve questions - Submit IBIS-ISS document to IBIS Open Forum for review -------------------------- Call for patent disclosure: - No one declared a patent. ------------- Review of ARs: - Arpad send AMI flow diagrams to Mike L. to be posted on the web - Done - Mike change web page to make items easily linkable - The pages no longer use frames - Next will try to make document URLs shorter - The minutes were broken for several months, but fixed now - Arpad Write a clarification BIRD to discuss accuracy issues related to the various AMI clock_tick algorithms in an IBIS-AMI DLL - TBD - Todd: Update the BIRD for IBIS S-parameter box based on feedback from discussion - 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: Vladimir gave a presentation on: "How to account for non-LTI of Tx analog buffer in IBIS AMI flow" which was followed by a questions and answers session. Walter thanked Vladimir for the presentation and thought it was excellent. Danil: What kind of non-linearities can be addressed by this, for example capacitors? Vladimir: Capacitors use a discretized model which is converted to difference equations which can be used in non-linear situations like for I-V curves. We just need to pay attention to how the model is partitioned to not have any double counting of the capacitance. Fangyi asked a question to which the response was: Equation 2 on slide 14 - model is like a general subcircuit, just a function call to get current. Fangyi: Non-linear questions are usually solved by Newton iteration. Where would this be done, in the Tx DLL model or the EDA tool? Vladimir: It can be either in the Tx DLL model, or in the "Utility DLL" (which would not have to be written by the model maker). Fangyi: Does this Utility DLL actually become a simulator? Vladimir: No. It is basically a function library. Fangyi: There are other types of methods, e.g. bi-section, to solve equations, and we also need controls for convergence, etc... Vladimir: These can be parameters to the function. There are many ways to solve equations. Fangyi: Who generates these function? Vladimir: Model vendor Fangyi: Was the model used in the presentation an IBIS model? Vladimir: The example was IBIS with I-V and V-t curves. It is easier to solve the equations in the presence of the admittance matrix than just solve the Tx output alone. Arpad: How important is this capability? Is it urgent to consider this to be included in the IBIS 5.1 specification? Fangyi: This situation is close to a three component channel model Tx-channel-Rx which is a general transient simulation Walter: We should not seriously consider this capability for the next IBIS version. IC vendors would not make models like this yet. Fangyi: Agree. Arpad: This is not like general transient simulation. Vladimir: This algorithm interacts with the Tx model's output only. Regarding model vendor not being able to write models like this, how do we know that the IBIS models currently used to generate impulse response are accurate enough? If we consider those accurate, this should be possible too. Arpad: Can the I-V and V-t curves of the IBIS model used for this capability? Vladimir: EDA can use those to make impulse response. Walter: Did you say that IBIS models can be accurate enough? Vladimir: Yes. Walter: Not true. IBIS models are not accurate to generate impulse response. Arpad: What is used to generate impulse response then if not IBIS? Walter: Touchstone broadband models. Vladimir: Cannot tell how big is the error. Walter: These are wonderful discussion points. Vladimir: Slide 12: Impulse response is not using Tx. It can be 100% accurate with Tx GetWave. Walter: Agree. Vladimir: The biggest obstacle is the ability to model Tx Walter: This is a significant task to model makers (IC vendors) Arpad: If we waited for the need to arise to include this capability in the IBIS specification, it may be too late, considering the time needed to get a BIRD written and approved, and get the new IBIS specification ratified. Walter: This doesn't have to wait years. Let's get the AMI corrections done first and then consider this. Ambrish: What are the complications considering cross talk? Vladimir: Cross talk is linear and a second order effect with respect to signal. Don't think this will be effected by that, like jitter isn't either. We only need one cross talk vector per channel. This proposal doesn't go into the details of the DLL interface, just wanted to show the concept of dealing with non-linearities. AR: Arpad send Vladimir's presentation to Mike L. for posting Next meeting: 20 Oct 2009 12:00pm PT -------- IBIS Interconnect SPICE Wish List: 1) Simulator directives