IBIS Macromodel Task Group Meeting date: 20 March 2012 Members (asterisk for those attending): Agilent: Fangyi Rao * Radek Biernacki Altera: * David Banas Ansys: Samuel Mertens * Dan Dvorscak Curtis Clark Arrow Electronics: Ian Dodd Cadence Design Systems: Terry Jernberg * Ambrish Varma Feras Al-Hawari Celsionix: Kellee Crisafulli Cisco Systems: Ashwin Vasudevan Syed Huq Ericsson: Anders Ekholm IBM: Greg Edlund Intel: Michael Mirmak LSI Logic: Wenyi Jin Maxim Integrated Products: * Mahbubul Bari Mentor Graphics: * John Angulo Zhen Mu * Arpad Muranyi Vladimir Dmitriev-Zdorov Micron Technology: Randy Wolff NetLogic Microsystems: Ryan Couts Nokia-Siemens Networks: * Eckhard Lenski QLogic Corp. * James Zhou Sigrity: Brad Brim Kumar Keshavan Ken Willis 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 The meeting was lead by Arpad Muranyi ------------------------------------------------------------------------ Opens: - None -------------------------- Call for patent disclosure: - None ------------- Review of ARs: - Mike post GetWave_Exists presentation to work archive - Done - Arpad send suggested GetWave_Exists wording changes to 5.1 editorial group - Done - Arpad to write a new revision of BIRD 117 and 118 to generalize references to parameters in files (.ami or any) - in progress - Ambrish update BIRD 145 for pad to pin mapping and other clarifications - Need to say what it will not solve - in progress ------------- New Discussion: Arpad: We need to know if [External Model] allows additional nodes BIRD 144: - Ambrish: Feras could not make it today, would prefer to defer - Arpad: The complexity of this originally simple proposal is growing - The value is becoming questionable - Ambrish: This could be discussed by email - Walter: Does BIRD 144 add something beyond what ISS has? - Ambrish: It eliminates the need for the ISS subcircuit - Some cases might be handled by direct association BIRD 145: - Arpad: We spent a whole meeting on this - Walter: The idea is to model all on-die circuits between a single buffer [Model] and pad? - A bank of 96 SerDes buffers can share the same power circuit - Ambrish: We should discuss this by email - James: We concluded that 5.0 already allows using [External Circuit] for power dist - Page 136 has an example circuit with a PDN - There are some pin mapping issues, but BIRD 145 does not resolve that - The diagram shows 2 buffers connected to 1 circuit, there could be more - Arpad: Agree, this BIRD introduces cascading models - Tools may have trouble telling which model in the circuit goes to a certain pin - Walter: A package model can model a slice of the package - You have a similar problem there too - Arpad: That is a more general problem that BIRD 145 need not address Arpad showed "The AMI analog model" from 25 Jan 2011: - Arpad: Are the isolation amplifiers needed? - Input signals are often from ideal sources anyway - Output signals usually have no load - Scott McMorrow had insisted on these - No recent supporting arguments - Walter: These are not needed - Mike: We should state that input sources and output loads must be ideal - Bob: The language could be made technically equivalent to the amplifiers - Walter: The simulator can add them only where needed - James: Something needs to terminate the left side block - Terry drew a circuit with a current driver and 50 ohm load - That is not backward isolated - Arpad: The EDA tool can figure this out - James: The question is what information to give the tool - Radek: This goes back to AMI methodology - The analog part is a single transfer function - We need a non-ideal source with a certain impedance - It can be communicated in the spec - Arpad: It could be included in an ISS subcircuit - Once a non-ideal source is specified you can't override it - Bob: Is the outlined box the s-param itself? - Radek: That could be other circuit types - Walter: If termination is needed it will be in the ISS subcircuit - Arpad: They can add linear gain elements too - James: Why have the unity gain amp then? - Arpad: To assure that there are no boundary condition requirements - James: A real IC could not have that - The signals on the left are AMI_Getwave output? - Walter: Yes - Radek: The amplifier after an ideal source adds no value - Bob: An s-param might have some input impedance - We can't leave it for the tool to calculate - Walter: It is the responsibility of the model maker to specify the input - James: It will be difficult to model Terry's circuit - Walter: The unity gain amplifier can be in the subcircuit - James: Can ISS model amplifiers with arbitrary impedance? - Walter: Yes - Arpad: Thevenin equivalents are easy - James: IBIS has not specified that - Arpad: We decided in Jan to have no analog model between Getwave and convolution - James: The flow shows Getwave connected to the analog model - Arpad: That is the impulse response - Walter: Engineers will come to the conclusion that it can be done that way - James: Why show the signals being input to the ISS block? - Arpad: We can take out these amplifiers and define the inputs and outputs - Bob: You don't know if you need the amplifier or not - James: Model makers will know when to put it in - The spec just needs to have everything that might be needed - Bob: An s-param block might not be interchangeable with an RC block - That needs to be spelled out - James: There is a question whether or not that needs to be spelled out - Bob: I will send a mathematical discussion to the group ------------- Next meeting: 27 Mar 2012 12:00pm PT Next agenda: 1) Task list item discussions ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives