IBIS Macromodel Task Group Meeting date: 17 September 2013 Members (asterisk for those attending): Agilent: Fangyi Rao * Radek Biernacki Altera: * David Banas Julia Liu Hazlina Ramly Andrew Joy Consulting: Andy Joy ANSYS: Samuel Mertens * Dan Dvorscak Curtis Clark Steve Pytel Luis Armenta Arrow Electronics: Ian Dodd Cadence Design Systems: Terry Jernberg Ambrish Varma Feras Al-Hawari * Brad Brim Kumar Keshavan Ken Willis Cavium Networks: Johann Nittmann Celsionix: Kellee Crisafulli Cisco Systems: Ashwin Vasudevan Syed Huq Ericsson: Anders Ekholm IBM: Greg Edlund Intel: Michael Mirmak Maxim Integrated Products: Mahbubul Bari Hassan Rafat Ron Olisar Mentor Graphics: * John Angulo Zhen Mu * Arpad Muranyi Vladimir Dmitriev-Zdorov Micron Technology: * Randy Wolff * Justin Butterfield NetLogic Microsystems: Ryan Couts Nokia-Siemens Networks: Eckhard Lenski QLogic Corp. James Zhou SiSoft: * Walter Katz * Todd Westerhoff Doug Burns * Mike LaBonte Snowbush IP: Marcus Van Ierssel ST Micro: Syed Sadeghi Teraspeed Consulting Group: Scott McMorrow * Bob Ross TI: Casey Morrison Alfred Chong Vitesse Semiconductor: Eric Sweetman Xilinx: Mustansir Fanaswalla Ray Anderson The meeting was led by Arpad Muranyi ------------------------------------------------------------------------ Opens: - Bob asked to give a brief BIRD 155 update. -------------------------- Call for patent disclosure: - None ------------- Review of ARs: - Arpad submit BIRD 155.1 draft 7 to open forum - Done. - Recommendation also made to reject BIRD 150. ------------- New Discussion: BIRD 155.2 update: Bob said he has been working with Radek to make some clean ups. All "DLL" have been changed to "executable model" for OS independence. A motion to vote will be made, as well as a motion to reject BIRD 150. Interconnect update: Arpad said interconnect meeting attendance has been low, and it was decided to suspend the meetings until further notice. Related discussion will take place in the ATM meetings. Package Modeling Decisions: Arpad showed and narrated a presentation. This contained a single slide overview of BIRD 125, BIRD 145, and EMD, plus a conclusion slide. Arpad noted that BIRD 125 lacks support for stacked die. BIRD 145 was deemed a quick, small specification change, and with "poetic license" could be used for package modeling. There was no recommendation however to do this. Arpad suggested that EMD might take over the role of defining the component from .ibs files, and it would require more work on the details. On the conclusion slide Arpad stated that it would not be good to have two approaches in IBIS. Walter made some clarifications regarding EMD, which stands for Electrical Module Description. He called it a replacement for the EBD Path syntax, using subcircuits. It is not intended as a package model syntax, but rather for modules with multiple components. His proposal for IBIS package models is different, and is based on data the industry has available. Walter showed a Package Model Requirement Survey slide from months ago. Arpad asked what Walter's package proposal would be, if not EMD. Walter clarified how EMD would not be good for package models, and he described the requirements for different package model types. David added that another type used is an s2p that adds a crosstalk path. Walter wondered if this should be added to the list of types. Slide 12 compared an "EMD-like" solution to BIRD 125. He said that the EMD-like proposal would work a dn that this probably could not be done using BIRD 145. John asked if the tool/user would have to make guided aggressor assignments. Walter said that the EDA tool would use coupling information to determine this. John asked if Touchstone 2 sparse format would help to represent coupling, where it could slide around and map to a larger s-param model. Walter said the sparse format can reduce file size, but it would be used to represent an s156p in s20p portions, for example. John said a script would be required. Brad said some tool would have to know how to create a coupled s156p from smaller pieces, progressing from uncoupled s2ps to coupled s6ps. Both known and unknown aggressors would be involved. John noted that the package model alone can not be used to choose all of the aggressors because it does not know enough about the board. Walter suggested that at times an s156p would be needed, but vendors might be hesitant to produce those. Users would have to use their influence. Brad noted that Scott McMorrow had said that unknown aggressors are not helpful, and that about 10 known aggressors are needed. He suggested that Scott and Walter only disagreed on the number of aggressors. Radek disagreed, saying that there can be indirect coupling, and that this is hard to predict by approximation. Brad agreed that building a model from smaller pieces constitutes an approximation. Walter said the suggested models would be difficult to produce using BIRD 125. Brad asked if Walter's idea would support 10 aggressor models. Walter said it would. Brad asked if having unknown or "maximum" aggressors would be useful, and how. Walter said an s156p could have maximum aggressors throughout, and that multiple package models for different purposes could be included. Brad asked how the tool would arrange s6p models to create larger models, the problem being unknown aggressors. Bob said Scott recommended using the s156p to determine the strong coupling areas. Arpad believed that this "building block" approach could be used with BIRD 125. AR: Arpad send Package Modeling Decisions presentation to Mike for posting ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives