=============================================================================== IBIS INTERCONNECT TASK GROUP Mailing list: ibis-interconnect@freelists.org ================================================================================ Attendees from July 24, 2024 Meeting (* means attended at least using audio) ANSYS Curtis Clark, Juliano Mologni Broadcom James Church Intel Corp. Michael Mirmak* [Michael Brownell] Xiaoning Ye Keysight Technologies Ming Yan Marvell Steve Parker MathWorks Walter Katz* Micron Technology Justin Butterfield Siemens EDA Weston Beal, Arpad Muranyi*, Randy Wolff* ST Microelectronics Aurora Sanna Synopsys Ted Mido, Edna Moreno Teraspeed Labs [Bob Ross] University of Illinois Jose Schutt-Aine Zuken USA Lance Wang* Michael Mirmak called the meeting to order. No patents were declared. During review of the July 17, 2024 meeting minutes, Arpad Muranyi noted a grammatical issue: "from the propose port location (XY) data" should be "from the proposed port location (XY) data" Arpad moved to approve the minutes with this change; Randy Wolff seconded. The minutes were approved with this change without objection. Michael noted that no new ARs had been assigned in the last meeting. Per an earlier AR to Michael, no new meeting series had been distributed to his knowledge, so recordings and transcriptions had not yet been enabled by SAE ITC. Randy confirmed this, adding that the last Open Forum was recorded, but no transcript was generated. Michael noted that his other ARs were still open. He added that he had been working offline on a means to use Git-like methods to collaborate on document updates. This would be discussed in a future meeting. Michael reviewed Draft 6 of the port-mapping proposal, supplied by Walter Katz on May 1 (labeled the "Y" draft). Michael asked whether the team was comfortable with using this document as a baseline for editing, and making "live" edits with comments directly in the document. Both Arpad and Walter agreed with this approach. Arpad noted that the tree structure concept for the proposal still needed to be discussed and decided. He added that he believed the current document to be a good start though less mature than needed; a lot of "loose ends" regarding syntax and technical issues still needed to be formally resolved. He observed that the format contains much more than just name/value pairs; this is also true of his proposal. Walter noted that he had another version of the document on his personal machine, but could use the Y draft 6 for editing in the group. Michael reviewed the basic structure of the document and proposed format. Walter noted that he had added Sij information for identifying data that was from measurement vs. simulation (SijStatus). Walter added that the proposal contains no tree structures; all data is in name/value pairs. Arpad replied that the structure of the proposal is OK; the name/value pairs should have pre-defined names unambiguously defined, however. In response to a comment from Michael, Walter noted that the Port argument value should be non-zero and less than the Number of Ports above in the file. Arpad asked whether reserved parameter names are also in the format of name/value pairs. Walter replied that the format is closer to Port # (Type ) (Physical ) (Logical ) etc. Reserved values or entries are listed in the proposal draft elsewhere. For C_comp Models, Randy noted that ExtRef may be missing. Walter replied that we have to verify that the C_comp Model listing is complete. [AR] Randy also noted that A_GND is a valid C_comp_model connection. C_comp_model is also one of the reserved sub-parameters for the keyword pair Begin/End Port Map. Arpad suggested that the format should use a keyword declaration approach. Randy noted that several structures, including those referring to IBIS structures, should be optional for transistor device models. Walter provided an over view of IBIS_Type, which takes two arguments; the first is a reserved word and the second is a filename. Randy noted that IBIS_file actually also takes Component as an argument. Walter stated that the format needs an additional set of parameters and arguments, including Layout for BRD file, ODB file, etc. CAD_Database and Schematic should be added as well. The language could resemble the following: The following optional subparameters identify what the Touchstone file port map is describing; only one of these subparameters is permitted per [Begin Port Map]/[End Port Map] keyword pair. The IBIS_File subparameter is optional. It is to be followed by an IBIS file name and component name. If present, this TS file represents an [Interconnect Model] that is in the IBIS file. EMD_File subparameter is optional. It is followed by an EMD file name. If present, this TS file represents an [EMD Model] that is in the EMD file. The C_comp_model_file model subparameter is optional. It is followed by a C_comp_model file name. If present, this TS file represents a C_comp_model. The Schematic keyword would be followed by a sheet number and node as arguments. Arpad asked whether the format should support connector, cable as arguments. Michael added that cable_assembly should also be supported, for completeness. Walter suggested this was covered under the Source keyword, including CAD_Database and Schematic. Arpad asked about Michael's proposed Expected_Passive parameter. Walter and Arpad suggested that the parameter should apply to the entire Touchstone file, not just one port. Michael proposed that this parameter be documented in a separate TSIRD. Walter agreed. SijStatus should be its own name/value pair. Some work is still needed on formatting. Walter added that the argument could be "VSS" rather than a type. He noted that Group is a list of pin names. Michael asked whether the team would edit this document if it were sent out with the meeting minutes. Walter agreed, stating that the information here is what we want to edit. What we store in this document is what we intend to pursue. He asked whether we intend to continue developing this format or Arpad's earlier proposed format. Arpad replied that, to squeeze all the requested information into his "HL" format proposal would take a lot of reformatting. This format (style, syntax) lends itself to including this new data. Walter stated that this format was directed toward getting people to read and understand the content, separate from machine readability. Arpad observed that he had originally been proposing his "HL" format approach due to its existence in his own firm's tools and its having been tested; some tweaks had been added to get new, needed information into the same syntax. The features being requested for the Y format draft may not be supportable in the HL format. Michael stated that the team needs to ensure the HL format content can be pulled into this Y format proposal completely, leaving nothing out. Arpad replied that this effort was not yet complete. Arpad moved to adjourn. Walter seconded. The meeting adjourned without objection. The next meeting will be held July 31, 2024. ================================================================================ Bin List: 1) Complete port naming proposal (Katz et al) 2) Complete/revise Touchstone 3.0 draft outline (Mirmak) 3) Complete ISS-IRD 1 Draft - enable cascading of S-parameters through W-element (Mirmak) - TABLED Tabled ARs: - Arpad to give an example of the physical connectivity needed for EMD automation.