IBIS Macromodel Task Group Meeting date: 17 December 2013 Members (asterisk for those attending): Agilent: * Fangyi Rao * Radek Biernacki Altera: David Banas Julia Liu Hazlina Ramly 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 * Andy Joy 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: - Arpad: We will have no meeting Jan 28, conflicts with DesignCon -------------------------- Call for patent disclosure: - None ------------- Review of ARs: ------------- New Discussion: EBD/EMD/interconnect direction: - Walter showed a presentation "IBIS-ISS Package Proposal Status" - slide 5: - Walter: We decided to support sNp models directly. - We need rules for a few "maybes". - slide 15: - Walter: We need to support sparse ports. - slide 17: - Walter: This will be presented at the upcoming summit. - This slide had been dropped but would like to discuss it there. - Bob: What does model type mean here? - Walter: It is about associating package models by [Model] Model_type - slide 20: - Walter: We may want distinct Delay_Corner and Xtalk_Corner. - slide 24: - Walter: I would like to resolve the first question here today. - We have not yet debated the actual proposed syntax. - The EMD syntax should be similar to package model syntax. - Bob: On slide 7 Keywords and Sub-Params are mixed up - Walter: Some editorial work is needed. - Bob: On slide 24 Model_name means [Model] name? - Walter: Yes - Bob: How to identify different signal paths with same model name? - Walter: There can be no joins, so they remain unique - Bob: How is a package model chosen when it might match multiple rules? - Walter: There are precedence rules. - Radek: This tries to save space in the file by introducing complicated rules. - That may not be a good idea. - Specifying each pin will not take that much space. - Walter: A package with 500 pins would need 500 entries. - These concepts have been requested by users. - Radek: Association by IBIS buffer model is good. - We still need to solve the problem of unused ports with coupled models. - John: Do we ever have 500 package paths that should be modeled the same? - Or will there be a variety? - Walter: Yes, either - We have had to bench measure packages to get all of the connection lengths. - John: There can be a problem of identifying victims but not distinguishing aggressors. - Walter: An s1000p can be used to do anything. - The crosstalk impact must be bounded to be practical. - This usually results in s12p models - Vendors are gaining technology to generate detailed models. - John: Touchstone2 sparse format would help - Walter: IC vendors don't want to generate those - They want a standard that sets a crosstalk limit, and that becomes the model. - We implement those bounded models in our tools, differently - Vendors would like to specify those directly - Walter: Can we decide that there will be direct Touchstone models? - Radek: It needs to be specified carefully. - Walter: We may fail, but I think we have decided to try. - Walter discussed slide 22, Unconnected Port Naming Rules. - Fangyi: Conventionally, any unconnected port is open. - The EDA tool doesn't have to do anything, the user will connect those nodes. - Walter: With 500 ports most will require some termination. - Fangyi: The EDA tool should handle that, not the model. - Radek: Then there will be different solutions in different tools. - We are trying to specify more of the netlist, outside of IBIS-ISS. - Walter: Vendors want to create 500 port models, knowing they will rarely all be used. - Fangyi: Tools can give users a selection of terminations. - Arpad: The model maker has an idea of what should be used. - Walter: There should be an override. - Fangyi: Agree, the user must have final control. - Walter: Yes, but a default should be given. - Arpad: We might want an option to use the Touchstone reference impedance. - Walter: That option is there, when R=0. - Fangyi: That is not a good idea. - Bob: On slide 20s, "p" and "n" would be better than "+" and "-". - Walter: I will ask for at least an hour to discuss this at the summit. AR: Walter send presentation to Mike for posting ------------- Next meeting: 14 January 2013 12:00pm PT ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives