From: ibis-editorial-bounce@freelists.org on behalf of Mirmak, Michael Sent: Thursday, May 14, 2015 6:04 PM To: ibis-editorial@freelists.org Subject: [ibis-editorial] Minutes, IBIS Editorial Task Group Meeting for April 24, 2015 ====================================================================== IBIS EDITORIAL TASK GROUP http://www.eda.org/ibis/interconnect_wip/ Mailing list: ibis-interconnect@freelists.org Archives at http://www.freelists.org/archive/ibis-interconn/ ====================================================================== Attendees from Apr. 24 Meeting Cadence Design Systems Bradley Brim Intel Corp. Michael Mirmak* Keysight Technologies Radek Biernacki* Mentor Graphics Arpad Muranyi* Micron Technology Justin Butterfield, Randy Wolff* Signal Integrity Software Walter Katz Teraspeed Labs Bob Ross* No patents were declared. Opens: Bob Ross will be out next Wednesday during the Interconnect Task Group meeting. Michael Mirmak reviewed the latest Draft 2 IBIS 6.1 changes. These include a move of Initial Delay in terms of placement in the document. Bob asked whether Initial Delay applies to [Submodel]. Randy Wolff added the question of whether it applies also to [Driver Schedule]. Radek Biernacki suggested the team needed to think about it. Bob noted that one can use Submodels for switched in/out models. He asked whether Initial Delay should be permitted under Submodel, particularly if the delays can be different. Radek replied that Initial Delay applies to all Rising/Falling Waveforms. Bob suggested that perhaps it applies to Submodel already, if additional text can be added to clarify this. Randy asked whether Initial Delay should be added to the Submodel tree? Bob responded only if it applies; we don’t have Composite Current under the Submodel. Is this another potential hole in the specification? Bob added that there were editorial problems in the hierarchy tree, including a gap in the linking characters above [Submodel] in the hierarchy, as well as another gap just above [Ramp]. Arpad Muranyi amount of overclocking could be different with independent numbers for rising and falling waveforms. Can/should we do this? Might end up being a BIRD. Bob: Dep in Usage definition, add “Illegal before Ver. 6.1”. Covers later usage. Radek: Isn’t that assumed for any new parameter? Bob: we are under a separate versioning scheme for .ami parameters here. Perhaps we should make a note that AMI_Version restricts model authors to using current keywords/parameters only. Need note in this section. Additional fixes needed 1) Bob: Fix tables and margins, starting with Table 20. Table 21 is missing Footnote 3. 2) Bob: Table 24 is missing Footnote 3. Same for Footnote 3 in Table 27. 3) Bob: “Dep” should use Footnote 4 in Table 30, plus a reference to AMI_Version 6.1. Model_Name and Resolve_Exists need to be added to Table 30, with Footnote 4. Fix margins on Table 30. Fix column sizes/widths on header (extra column on Table 30 and 24; check all tables for this and Allowable Usage). 4) Bob: Model_Name and Resolve_Exists need to be added to Tables 31 and 32, with Footnote 4. 5) Radek: four statements include the phrase “a the” – should be “the” in all four statements. 6) Radek: matrices are not reciprocal or passive; they can be symmetric, however. Michael suggested that the sentence “Package RLC matrices are assumed to be reciprocal and passive…” should start instead with “The circuits described by package RLC matrices…” 7) Michael: p. 179 uses “touchstone” instead of “Touchstone”. 8) Radek: on resolving dependences, the example on p. 179 should note that the examples use concepts not yet defined in the specification until later in the document. 9) Michael: fix quotation mark format in the example on p. 179. Also “parameter” is misspelled as “paramter”. 10) Michael: re-add the phrasing showing “infinite”, replacing with “unlimited” for Dep section. Examples in original phrasing were for applications, not specific syntax. 11) Bob: example should use indents on Descriptions, to move “AMI_Resolve” and “OP_mode” inward. Better to change examples to avoid line spanning. Michael also noted that there is inconsistent usage of “must” and “shall” in the document. He clarified that “must” implies action by specific people, while “shall” implies definition. At the next meeting, a new draft will be discussed and Initial Delay and specific usage of Rising/Falling Waveforms will be examined. The suggested fix is that Initial delay apply to top-level models only for Submodel.