================================================================================ IBIS INTERCONNECT TASK GROUP Mailing list: ibis-interconnect@freelists.org ================================================================================ Attendees from February 7, 2024 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 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. Weston Beal made some comments regarding the use of the hyphen in "S-parameters" where "S" is used as a modifier of "parameters". Michael reviewed the minutes of the January 24, 2024 meeting. Arpad Muranyi moved to approve the minutes; Weston seconded. The minutes were approved without objection. During the AR review, the team briefly discussed an older AR given to Michael regarding reviewing IEEE 370 and sampling. The language of the standard is complex; Michael will continue reviewing the text and report on his findings in a future meeting. Arpad reviewed the latest draft of the pole-residue proposal. Draft 20 changes from Draft 19 include green sections (added) and removal of yellow highlighted areas. Green text adds the key details about requirements (rules) and both the reasons for the data described and how it is derived (min valid frequency, for example); some material was suggested by Randy Wolff. Support for upper/lower frequency matrices is causing issues. The parser will need more complicated logic to handle this. The Number of Pole-Reside Indices keyword can include multiple index components. A new example has now been included in the text, to cover two matrix pairs being listed, which implies identical data. The format now counts indicies rather than keywords. Weston suggested changes to the Begin Pole-Residue section to simplify it. One now has as many keywords as index pairs, but one could have less (with shorthand notation). The key concept is that coverage of all the pairs is needed across as many or as few keywords as needed. The use of Matrix Format will change the number of indices to be included, according to Arpad. Michael asked whether adding this support is worth the effort. Arpad noted that this would result in a larger specification and a more complex parser. But the support text is there; Bob Ross requested this support. Weston replied that the parser is not that much affected by the addition. Missing pairs are explicitly stated to be zero in this draft. Walter Katz asked whether the parser will generate full matrix data for all i,j for sparse, non-full, etc. arrangements. Weston replied that the Touchstone 2.1 parser will have to. With the ability to change from Touchstone 2.0 to 1.0 files, it must have all the data filled in. Michael related the recent request from a DesignCon conversation with Pete Pupalaikis, who requested having parser data structures available in memory or as a separate file even in the free syntax-checking executable. Michael accepted the AR to contact the Quality Task Group for details. [AR] Arpad noted that the Begin Residues Data keyword is a duplicate of Begin Poles. He asked whether reciprocal data is being handled. Weston suggested it was already being handled. He added that he had never heard of a reciprocal matrix in math class. Walter replied that a reciprocal network is where Sji = Sij (particularly for power networks). This is available in most references. Weston suggested throwing out reciprocal mentions and just talking about symmetrical matrices. Michael asked whether the team had reached a decision yet: support the Matrix Format, support the indexing method, support both? The general consensus was to support just Matrix Format. Walter suggested leaving the text the way it is. Even binary compression only gives you a factor of 2 reduction in size anyway. Weston suggested that the keyword usage should be Matrix element not Matrix component. Walter stated that tools may read the matrices into SPICE elements; every simulator may do that differently, so results may differ. Weston noted that this is similar to IBIS. Walter agreed, but the result differences are much less with IBIS (except in handling of K-t tables and C_comp). Michael mentioned a need for a complete example. Otherwise the document appears very close to being done. Arpad moved to adjourn the meeting; Weston seconded the motion. The meeting adjourned. The next meeting will be held on February 14, 2024. ================================================================================ 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.