IBIS Macromodel Task Group Meeting date: 15 November 2022 Members (asterisk for those attending): Achronix Semiconductor: Hansel Dsilva Amazon: John Yan ANSYS: * Curtis Clark * Wei-hsing Huang Aurora Systems: * Dian Yang Cadence Design Systems: Ambrish Varma Jared James Google: Hanfeng Wang GaWon Kim Intel: Michael Mirmak * Kinger Cai Chi-te Chen Alaeddin Aydiner Keysight Technologies: Fangyi Rao Majid Ahadi Dolatsara Ming Yan Radek Biernacki Rui Yang Luminous Computing David Banas Marvell Steve Parker Mathworks (SiSoft): Walter Katz Mike LaBonte Micron Technology: * Randy Wolff Justin Butterfield Missouri S&T Chulsoon Hwang Yifan Ding Rivos Yansheng Wang SAE ITC Michael McNair Siemens EDA (Mentor): * Arpad Muranyi Teraspeed Labs: * Bob Ross Waymo: Zhiping Yang Zuken USA: * Lance Wang The meeting was led by Arpad Muranyi. Curtis Clark took the minutes. -------------------------------------------------------------------------------- Opens: - None. ------------- Review of ARs: - Arpad to submit the latest clock_times clarification BIRD draft to the IBIS Open Forum as an official BIRD. - Done. Submitted as BIRD222. -------------------------- Call for patent disclosure: - None. ------------------------- Review of Meeting Minutes: Arpad asked for any comments or corrections to the minutes of the November 8th meeting. Randy moved to approve the minutes. Lance seconded the motion. There were no objections. -------------- New Discussion: Standard Power Integrity Model (SPIM) BIRD draft: Kinger reviewed the latest draft of the proposal. He reported that he had incorporated the fix Arpad and Bob had suggested at the last meeting. The File_IBIS-ISS Sub-Param of [SPIM Rnetwork File] now specifies the subcircuit name as well as the file name, as is the case for existing usages of File_IBIS-ISS in IBIS 7.1. Randy observed that the example had been corrected, but he said the Usage Rules for [SPIM Rnetwork File] would also have to be updated to reflect the change. Bob asked whether it was correct to have [SPIM Voltage List] under [SPIM Rnetwork File] in the hierarchy. Kinger said it was. He said [SPIM Voltage List] is provided for DC IR drop simulations only, and it is properly scoped inside [SPIM Rnetwork File]. It is not needed for AC simulations with [SPIM Touchstone File], for example. The group provided other editorial feedback, including: - File_IBIS_ISS -> File_IBIS-ISS - S parameters -> S-parameters - [End SPIM Chip Group] -> [End Chip SPIM Group] - Remove "Intel" from the [Manufacturer] Usage Rules example Kinger said he would update the draft with the changes discussed in the meeting and send the new version out to the ATM list for further review. PSIJ Sensitivity BIRD: Kinger noted that Michael Mirmak had been added to the Requestor list. He invited anyone else interested in advancing this proposal to join the list. Kinger reviewed changes in the latest version, draft2. He said the majority of the changes relative to draft1 were in the Introduction section. Derivation details, equations, and figures from recent IBIS Summit presentations are included in the Introduction. In the Solution Requirements section, a new Table of IBIS Keywords and Sub-Params had been added. Kinger said [PSIJ Sensitivity Group] is analogous to [Model Selector]. A single DDR controller might work for multiple standards, or a single PHY might support multiple protocols. The power rails might be the same, but PSIJ sensitivity would be different for each different protocol or standard. Each PSIJ Sensitivity Group would specify all the power rails for the I/O interface operating in a particular mode. For the [PSIJ Sensitivity] keyword, Kinger reviewed the example and the accompanying Figure 2. He said we need to describe how the model maker should choose the frequency data points to provide in the table. He said the goal is for the model maker to provide the worst-case envelope for PSIJ sensitivity, not to duplicate the fine structure in the data. Randy noted that the example's figure seemed to indicate a spline or something other than a simple PWL was used to fit the data provided in the table. He suggested the BIRD should provide more information if a specific type of fitting is expected from the EDA tool. Arpad noted that one example used comma as the thousands separator character (e.g., 10,000), and he said he wasn't sure if IBIS supported that format for the numbers. He suggested Kinger use scientific notation in the example. Bob noted the use of [Voltage List] in this BIRD might be problematic because it appeared in a different scope than the existing [Voltage List] in IBIS 7.1. Bob said that in EMD the [Voltage List] is completely optional, and the model maker can decide whether or not to include any particular rail. Kinger said [Voltage List] is required in this BIRD, and it must contain the name of any power rail listed inside any PSIJ Sensitivity Group. He noted, however, that it is legal to have rails specified in [Voltage List] that do not appear in any PSIJ Sensitivity Group. He said a particular rail might be negligible for PSIJ sensitivity, or it might not appear in a particular PSIJ Sensitivity Group. Since the usages are different, Kinger asked whether we could resolve the issue by proposing a new keyword name, [PSIJ Voltage List], with this BIRD. Bob said this might be the solution, but we have to think about it. Kinger said he would create a new version incorporating the changes discussed in the meeting and send it to the ATM list for further review. - Curtis: Motion to adjourn. - Randy: Second. - Arpad: Thank you all for joining. AR: Kinger to send an update (draft5) of the SPIM BIRD to the ATM list. AR: Kinger to send an update (draft3) of the PSIJ Sensitivity BIRD to the ATM list. ------------- Next meeting: 22 November 2022 12:00pm PT ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives