IBIS Macromodel Task Group Meeting date: 03 April 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: - 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: BIRDs 117/118: - Arpad: An email was sent to the reflector to encourage discussion - Arpad showed BIRD 117.3 - Arpad: This shows new [Begin Parameter Tree] and [End Parameter Tree] keywords - Should we allow multiple trees in one keyword? - Mike: We should be able to allow multiple trees - Bob: We only allow one entity for each [Define Package Model] - Arpad: [Define Package Model] defines the name, this is different - Radek: The singular word "Tree" in the keyword implies only one - Arpad: We can make it "Trees" - Fangyi: The rules for this would be different than for AMI - Arpad: Yes the BNF would be a little different - The AMI rules are in section 10 - BIRD 127 gives more information - Bob: Is there a problem with reserved words like Vinh? - Arpad: There should be no problem - Bob: It may be impractical to put this in the IBIS file - The syntax is different yet it would have to follow IBIS rules - Lines would still be limited to 120 characters - Parser complexity may be a problem, and we have limited parser funding - Arpad: The parsing and syntax checking are separate issues - Mike: The parsing of data in nested parentheses should be separate from the checking for Reserved_Parameters, etc. - Walter: We could have a concept of .ibs files being containers of IBIS plus other file data - It would go after the [End] - No TreeRootName would be needed - Arpad: Then there could be only one tree - Ambrish: We could give the new keywords name values to be used as virtual file names - Arpad: Then the tree name would not be used AR: Arpad proceed with BIRD 117 change to allow multiple trees Analog BIRDs 116 & 144: - Arpad: We should close on the question of if ideal voltage sources are needed - James: It would be best if we did not define the contents of the models explicitly - Arpad: Can we agree that it is sufficient for the specification to define the ideal sources which drive the Tx model or probe the Rx model? - Fangyi: I think so - Bob: D2A converters are needed for any analog interface, and these are ideal sources - Arpad: We should be able to specify that these sources must be present at the interface - Walter: At the output of the RX we still have an analog signal - Walter: They are required to know the port order - Arpad: We have to make it clear that there must be no load at the RX - Fangyi: Why do we call it an ideal voltage source at the RX? - It is really a probe - Arpad: Agree - Walter: For AMI it is an A2D that outputs a floating point number AR: Arpad propose IBIS spec changes to clarify ISS D2A & A2D interfaces - Bob: At some point we can't just edit the BIRD - Walter: The RX A2D samples at intervals to produce a series of numbers - Bob: TTL defined an X region between two voltages - Walter: The IBIS RX output has to always be effectively analog - A D2A for the TX makes sense - Arpad: We at least agree the isolation elements are not needed in the subcircuits - James: How does the TX D2A handle the range of voltages needed? - Walter: The voltage swing is defined - Arpad: This is in the spec - The TX D2A input uses true/false logic - James: How is the AMI waveform applied to that input? - Fangyi: It is convolved with a response - It can be even a square wave, -0.5V to +0.5V, jittered BIRD 144.3: - Arpad: There is a question whether the BIRD remains worthwhile with extra complications - Feras: The main change is referring to a Touchstone file instead of SPICE - The other changes are for port terminations - The complexity is not much more than A2D and D2A statements - It forces users to generate correct circuits - Bob: Calling Touchstone directly makes sense, it should not be banned - Touchstone is more general, for PDS and coupling applications for example - IBIS-ISS can serve those needs too - Port termination is an EDA tool issue, not a model issue - The tool should be able to figure out the Touchstone file - I don't support the port termination language - Arpad: How would the model maker convey how to connect the Touchstone model? - Bob: That is an EDA tool issue - Radek: You can't avoid specifying the whole circuit - Bob: The tool might want to cycle through all possible combinations - Feras: The tool should not have to guess what to do with extra ports - Ambrish: Circuits should be provided that completely describe the connections - We should provide the simplest possible description - Arpad: Will the model maker know if the D2A is ideal, and a resistor must be added ------------- Next meeting: 10 Apr 2012 12:00pm PT Next agenda: 1) Task list item discussions ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives