IBIS Macromodel Task Group Meeting date: 01 March 2016 Members (asterisk for those attending): ANSYS: * Dan Dvorscak * Curtis Clark Broadcom (Avago): Xingdong Dai * Bob Miller Cadence Design Systems: * Ambrish Varma Brad Brim Kumar Keshavan Ken Willis Cisco: Seungyong (Brian) Baek eASIC: David Banas Marc Kowalski Ericsson: Anders Ekholm GlobalFoundries: * Steve Parker Intel: Michael Mirmak Keysight Technologies: Fangyi Rao * Radek Biernacki Maxim Integrated Products: Hassan Rafat Mentor Graphics: John Angulo * Arpad Muranyi Micron Technology: * Randy Wolff Justin Butterfield QLogic Corp.: James Zhou Andy Joy SiSoft: * Walter Katz Todd Westerhoff Mike LaBonte Synopsys: Rita Horner Teraspeed Consulting Group: Scott McMorrow Teraspeed Labs: * Bob Ross TI: Alfred Chong The meeting was led by Arpad Muranyi. -------------------------------------------------------------------------------- Opens: - None. -------------------------- Call for patent disclosure: - None. ------------- Review of ARs: - Arpad noted that he had produced a second draft of his MinMax BIRD aimed at clarifying how missing min/max data should be handled. - Done. ------------------------- Review of Meeting Minutes: - Arpad: Does anyone have any comments or corrections? [none] - Arpad: Motion to approve the minutes. - Bob R.: Second. - Arpad: Anyone opposed? [none] ------------- New Discussion: How to handle missing min/max data?: - Arpad: [Sharing draft 2 of the MinMax BIRD] - Discussion: Arpad noted that he had simply taken draft 1, which had been written with respect to IBIS 6.0, and updated the references to IBIS 6.1. Draft 1 itself had already contained the list of all the places in the spec that referred to typ/min/max data. Arpad noted inconsistent language and the fact that some locations explicitly stated that typ should serve as the default for missing min/max data, while other locations said nothing about what to do if min/max columns were missing ("NA"). He said he had started this topic originally because there were many interdependencies amongst various parameters, and no rule was spelled out with respect to min/max data. He said that without a consistent set of rules being stated we left open the door to many typ/min/max combinations that were actually illegal or non- sensical. Arpad noted that this topic had recently come up in the context of Walter's proposed view that IBIS is a "measurement spec" and therefore should not specify how the data in a model is used. Arpad had used this typ/min/max language as a counter example, and now stated that if we took Walter's view we would not want to specify what to do if min/max data were missing. He said we needed to be consistent regardless of which direction we chose. Bob R. suggested that we might provide a general advisory statement that if certain parameters, subparameters, etc. were missing data in a column then simulations for that corner might not be valid. He said that he did not want us to start checking for this consistency in the parser, as the combinations were limitless. Radek agreed, but noted that there were likely hidden rules enforced by the parser, for example matching v(t) waveform steady state endpoints to I/V data would preclude corner cases if all three columns were not provided for all data. Bob R. confirmed Radek's statement and said the parser did enforce many relationships. Bob R. and Radek again expressed their disagreement with the "measurement spec" interpretation. De-embedding C_comp IBIS-ISS Subcircuits: - Discussion: Walter gave a brief follow up to last week's presentation. He wanted to clarify one idea he felt he might not have explained clearly. - If a [Model] has a C_comp and a C_comp subcircuit, then the EDA tool would use the C_comp to do the de-embedding (derive the K(t)), but not use it in simulation. In simulation it would use the C_comp subcircuit. If a model maker used a tool that provided this capability, it would be easy for the model maker to sweep the C_comp values and find the one that yielded the best match to the actual waveforms. - Walter: Randy and I will be updating the C_Comp BIRD to reflect this idea. - Arpad: Thank you all for joining. ------------- Next meeting: 08 March 2016 12:00pm PT ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives