================================================================================ IBIS INTERCONNECT TASK GROUP https://urldefense.proofpoint.com/v2/url?u=http-3A__www.ibis.org_interconnect-5Fwip_&d=DwIGAg&c=nKjWec2b6R0mOyPaz7xtfQ&r=DcQR-qLpQg5lIreuM6-NYECRIAFXt268PRNS5WO043M&m=Ht-RjhtOMbXp06ks5ANg5JvPkOas6GQ-tvxIceZ2dEnSGrHLxs8zlSshlix2Og6f&s=9sBy1RLNxNcvbUghnS9z3-Rr6W9nyHsouLyS7TJXez8&e= Mailing list: ibis-interconnect@freelists.org Archives at https://urldefense.proofpoint.com/v2/url?u=http-3A__www.freelists.org_archive_ibis-2Dinterconn_&d=DwIGAg&c=nKjWec2b6R0mOyPaz7xtfQ&r=DcQR-qLpQg5lIreuM6-NYECRIAFXt268PRNS5WO043M&m=Ht-RjhtOMbXp06ks5ANg5JvPkOas6GQ-tvxIceZ2dEnSGrHLxs8zlSshlix2Og6f&s=xn-WtfzDLhiLFtyiflu0pWd0YjE0mnVqMFTYvniw87g&e= ================================================================================ Attendees from May 25, 2022 Meeting (* means attended at least using audio) ANSYS Curtis Clark Intel Corp. Michael Mirmak Michael Brownell Keysight Technologies Ming Yan Marvell Steve Parker MathWorks Walter Katz Micron Technology Justin Butterfield* Randy Wolff* Siemens EDA Arpad Muranyi* ST Microelectronics Aurora Sanna Teraspeed Labs Bob Ross* University of Illinois Jose Schutt-Aine Zuken USA Lance Wang* Randy Wolff convened the meeting. No patents were declared. Justin Butterfield took minutes. Review of Minutes: - Randy called for review of the minutes from the May 18, 2022 meeting. Randy displayed the minutes. Randy noted the date of the meeting was not correct. Bob Ross motioned to approve the minutes with the date corrected. Lance Wang seconded. The minutes were approved without objection. Review of ARs: - Arpad Muranyi to give an example of the physical connectivity needed for EMD automation [AR]. - Randy suggested to keep this open. - Walter to send the draft PLS document. - Done. Opens: - None. Discussion: PLS in Touchstone TSIRD Draft: Arpad noted that he has not reviewed the PLS document from Walter yet. Bob stated that we have to decided how to partition the PLS and port naming. He suggested to have separate sections. Randy asked if this means separate sections in the specification or in the Touchstone file. Bob replied potentially both in the specification and in the file, but mostly, he is concerned with the specification for now. We still need to decided how to structure the Touchstone files. Arpad asked about the duplication of keywords in the specification. Bob is in favor of having some duplication. Arpad suggested to not duplicate the port naming. Bob stated the port naming could be a separate section that can be used for both Touchstone and PLS data. Randy stated each keyword is a new section. Arpad commented some keywords would be duplicate, but he does not like that idea. Bob noted we would not want to repeat keywords if the information should be the same. Bob noted an additional section could be the header section, which could be common for PLS and Touchstone. The port naming could be in the front of the file. Randy noted the file can be read more easily if the PLS data is before the network data. Arpad noted the biggest question is what to do with the option line. We kept it in Touchstone 2 to help the EDA tools with backwards compatibility. Randy noted you still have to down convert a Touchstone 2 file to use the file in a Touchstone 1 only tool. Arpad commented, if we have the PLS data, it would have to be converted with a frequency sweep back to network data. Bob stated the option line is not something we can remove from Touchstone, since there are too many tools that use it. Randy asked about the option line for the PLS data. Bob suggested to keep the option line and network data as is. Randy asked if any of the TSIRDs improve the network data. Bob stated some of them do consider the network data and some would apply to PLS, but we would have to think about this. Arpad suggested we could move the option line next to the network data and not have it for the PLS format. Randy suggested we need to describe an order for the keywords. Bob stated we need blocks of syntax. We could have blocks of keywords for network data, PLS, port naming, and header information. We would need to make the rules about the order of these sections and which ones are optional. There could be upward compatibility issues if we are not careful with these rules. Bob asked if we can mix Touchstone versions with PLS. Arpad replied they can support Touchstone 1 and Touchstone 2 conversions to PLS. Bob noted this would be another rule if we wanted to enforce it. Arpad asked if we want to complicate the Touchstone 3 specification by including the legacy features and related rules. Bob suggested we should not deprecate the Touchstone 1 format that everyone uses. Arpad asked about mixing versions for different sections. Randy thought having different versions for each section would not work well. If you add the Touchstone 3 features, then it is a Touchstone 3 file. Randy noted we do not do this in IBIS files. Arpad noted the AMI and IBIS could be different versions. Randy noted they are separate files. Arpad asked, if we have the port naming and PLS data making the model a Touchstone 3 file, why do we need to keep the old features. Randy replied that the tool parser should support all the Touchstone 3 features. He noted the parser could down convert from Touchstone 3 to older versions. Lance asked if we are considering keeping the PLS and network data in the same file. Randy replied that we are considering this. Lance asked if the data could be different between the two formats. Randy replied that this would be an egregious model error, but he was not sure that it could be checked. Bob stated we cannot check the content, but we can check the number of ports. Arpad asked if the down converter could move the port naming to comments. Randy and Bob agreed this would be a good idea. Randy noted you cannot convert the Touchstone 2 to Touchstone 1 if you have the per-port reference impedance. Bob agreed there are conversion limitations. Randy commented that the option line could be handled to move between versions. Randy commented the option line does not work well for the PLS data. Bob suggested to block out groups of information in the syntax. He stated it will complicate the rules to remove the option line. Arpad noted that having the option line complicates the rules since we have duplications and would have to check both. Randy suggested to write down some the rules to start looking at the issues. Arpad noted, currently, we can have a discrepancy between the option line and the per-port reference impedance. Randy asked if the per-port reference can support a single value. Arpad replied all reference impedances must be listed. Arpad noted we could have more complicated rules and options for defining the reference impedance. He also suggested we should consider complex numbers in the Reference keyword. Arpad motioned to adjourn. Bob seconded. Next Meeting: The next meeting will be June 1. ================================================================================ Bin List: 1. Touchstone 3 2. Pole-residue support for Touchstone 3. Port naming