=============================================================================== IBIS INTERCONNECT TASK GROUP Mailing list: ibis-interconnect@freelists.org ================================================================================ Attendees from October 2, 2024 Meeting (* means attended at least using audio) ANSYS Curtis Clark, Juliano Mologni Arista Networks Jim Antonellis* Broadcom James Church Intel Corp. Michael Mirmak*, 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 University of Illinois Jose Schutt-Aine Zuken USA Lance Wang Michael Mirmak called the meeting to order. No patents were declared. Arpad Muranyi noted that he recommended three changes to the minutes as distributed: 1) From the original minutes: “Arpad noted that there are several reference nodes defined in the first example, which cannot be connected in EMD, as it only supports "per interface referencing".” The minutes should say something like this instead: “Arpad noted that there are multiple reference nodes defined in the first example, one for each interface, which cannot be connected in EMD, as it only supports a single reference connection for the entire S-parameter model.” 2) From the original minutes: “Walter Katz that added” Michael noted that this should be "then added". 3) Should the word "references" below instead be “examples”?: “The first 5-6 references of the text” Arpad moved to approve the minutes with these changes. Randy Wolff seconded. The minutes were approved without objection. The team agreed that referencing would not be discussed until an EM expert could be found to either comment on the port mapping draft or participate directly in the meeting. Arpad raised several individual concerns about the port mapping proposal. One is that the Net name on the negative side of a differential port may be needed; this is unrelated to referencing. Walter replied that the purpose of Net name is to specify which nodes are "connected". Arpad responded that this implies use of a unique Net name for each node in the pair, not a single net name for both. Michael referred to the strings DQS_t and DQS_c - aren't they already separate? Walter stated that the discussion is about reference net names; for differential ports, VNA measurements are never made with a probe on the high side and gnd on the reference side. Differentials can be measured on a single terminal with a reference; some are conducted with four probing nodes (two "references", one for each single-ended side). Two "ports" can form two differential pairs, four nodes at each end. Michael asked about an identifier for user-defined parameters. In the absence of one, should there be a separate section for such parameters. Randy suggested this could be organized similar to AMI. Walter replied that these are associated with ports, so no section separation would make sense. He suggested the prefix "UD:" for such parameters. Arpad agreed with Walter's comment, noting that this information should be on independent lines. Arpad asked whether these parameters should be on the same Port line. Walter replied that these should be in-line, name-value pairs only. Arpad asked about multiple arguments/values. Walter stated that each declaration can use multiple colon characters or multiple parameter declarations. Michael asked whether more parentheses pairs would be needed. Walter stated that these should be kept on a single line. Arpad took the AR to check on Ext_Ref spelling as used for IBIS Interconnect. He quickly closed this, confirming that the IBIS spelling is "Ext_ref". For EMD both Walter and Randy agreed that only pins or nodes are needed. Many different options are available. Walter stated that he believes we can make the reference node a bus_label. Randy accepted the AR to check on this. [AR] Michael noted that the draft still needs examples after Example 8 to cover non-Port subparameters. Walter noted that, for Physical parameters, the colon character is a separator. This identifies when is the argument is a refdes pin number vs. an xyz location. Arpad replied that the draft does not define syntax for xyz, and the draft should define units as well. Walter suggested that units be declared independently, with no mixing of unit declarations in the Port declarations. Arpad asked how the proposal will distinguish between string names and names with specific meanings. Arpad moved to adjourn; Walter seconded. The meeting adjourned without objection. The next meeting will be held October 9, 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.