================================================================================ IBIS INTERCONNECT TASK GROUP Mailing list: ibis-interconnect@freelists.org ================================================================================ Attendees from September 27, 2023 Meeting (* means attended at least using audio) ANSYS Curtis Clark Broadcom James Church Intel Corp. Michael Mirmak* Michael Brownell Keysight Technologies Ming Yan Marvell Steve Parker MathWorks Walter Katz Micron Technology Justin Butterfield Siemens EDA Arpad Muranyi*, Randy Wolff* ST Microelectronics Aurora Sanna Teraspeed Labs Bob Ross* University of Illinois Jose Schutt-Aine Zuken USA Lance Wang* Michael Mirmak called meeting to order and took minutes. No patents were declared. During the review of the previous meeting minutes, Arpad Muranyi observed that the text contained the sentence, "Michael clarified that the presentation S-parameters [sic] to normalized Z-parameters." The phrase "covers the conversion of" is missing from the middle of the sentence. Arpad also noted that his current workload included not only pole-residue improvements but also the port-mapping proposal. Arpad moved to approve the minutes with these changes; Randy Wolff seconded the motion. No objections were raised and the minutes were approved. Michael noted that his ARs to contact Sam Chitwood and Pete Pupalaikis are still active. During Opens, Arpad stated that he added many different examples to the port-mapping proposal, as part of an Interconnect reflector the previous Saturday. Bob Ross stated that he has a presentation on matrix parameters and Touchstone. He would like technical review in the next meeting. Arpad mentioned that he would be giving a joint presentation with Weston Beal at the IEEE EMC SI/PI IBIS Summit. He asked whether this should be repeated at other summits. Automation and EMD model-making were topics; Bob added that the upcoming Shanghai IBIS Summit will only be held in-person and would not be a hybrid event. Michael reviewed several questions he had regarding development of the Touchstone 2.1 draft document. Arpad asked about how TSIRD 4.1 was to be integrated. Michael explained that 4.1 and 4.0 were technically identical; only some documentation outside the new specification text was added in 4.1. Michael observed that there was inconsistent mention of Touchstone 2.1 vs. 2.0 in the TSIRDs. Arpad replied that version 2.0 does appear often, but not in headings; the document itself is being revised to 2.1 but not the syntax. Randy pointed out that specific Version 2.0 and 2.1 text appears in the new Version keyword definition text. Arpad suggested that the phrase "Rules for Version 2.0 and 2.1" would be acceptable. Randy provided the alternative, "Verison 2.0/2.1". Bob suggested using "and" or "or" as opposed to the "/" character. Michael proposed putting the phrase, "Note that except for the additional argument" in bold for emphasis. He will provide a draft to the reflector before the next meeting. Arpad reviewed his new port-mapping enhanced syntax. New blue-highlighted text has been added, such as "pin rail" with "pin name" or with "pad name" but not a buffer rail. Some text has been added on per-interface referencing; slides 11 and 18 use the same example where devices have their own GNDs as pin groups. He noted that EMD references prohibit more than one reference terminal. The Open Form may want to address this in EMD. Michael asked whether the presentation is intended to be a separate document, part of Touchstone, or a cautionary presentation just to ensure we don't have potential holes. Arpad's IEEE EMC presentation was intended as a warning: "you can make bad EMD models; don't do it". Michael asked whether we need to get rid of ground in IBIS to support Touchstone improvements. Arpad replied that this would not be neeed. The rest of the presentation documents potential EMD holes/conflicts. Two examples show packages with pin grouping, including pin IO, pad IO, pin rail, pad rail, plus pin to pad, and pad to buffer terminal connections. Arpad added that DESCRIPTION and pin descriptions should be mutually exclusive. Several strings in the examples, including PinSig or Side: Tx (the Tx portion) are optional. Bob asked whether the comments shown would be removed in the official version. Arpad confirmed this. Bob added that he believed everything in the proposal should be optional, as Walter Katz described. Arpad noted that he generally agreed, but this may not be possible in all cases; some things should be mutually exclusive, and some should follow the IBIS rules as noted for terminal types, etc. Bob replied that some of these elements are redundant. Arpad asked whether several features were needed if EMD already includes them. He has a specific flow of events in mind as an example: start with a board design, extract information from thre board with designator names, pad names, etc. The header or map will tell us how we can take the resulting Touchstone file and connect it to other elements. Arpad noted that he is assuming EMD is created using the Touchstone file, not the other way around. Bob replied does not want the Touchstone file syntax to cross-reference an EMD for validity. Michael asked whether the Touchstone file should be completely self-contained. Bob replied that he was not sure; contradictory information should be thrown out. Bob stated that the EDA vendor, not the user, should provide the mapping. Randy observed that tools will do this, not users; this is too complex an expectation for users. Michael added that we need this mapping for tools, plus four examples: - two-ended cable, with NEXT and FEXT - two-ended cable, differential - RF transistor - three-ended cable Bob added that he has a pathological 12-port global ground example, with every port connected with a 50-ohm resistor. Randy observed that the NEXT/FEXT aggressors case is interesting. For the next meeting, the agenda will include review of Bob's matrix math technical presentation. Arpad move to adjourn the meeting. Randy seconded the motion. The meeting adjoured. The next meeting will be held October 4, 2023. ================================================================================ Bin List: 1) [Complete draft Touchstone document separating version 1.0 and 2.0 into their own chapters] - REMOVED 2) Create structures to encapsulate Touchstone 1.0 data in Touchstone 2+ specifications - TABLED 3) Complete draft Touchstone 2.0 document containing TSIRD3 and TSIRD4 draft (Muranyi) – COMPLETED IN DRAFT 10 4) Complete pole-residue format BIRD draft (Muranyi) 5) Complete port naming proposal (Katz) 6) Create alternatives to the Touchstone 1.0 option line before the "R" character - TABLED 7) Complete ISS-IRD 1 Draft - Enable Cascading of S-parameters Through W-element (Mirmak) - TABLED 8) Complete/revise Touchstone 3.0 draft outline (Mirmak) – dependent on several items above Tabled ARs: - Arpad to give an example of the physical connectivity needed for EMD automation.