====================================================================== IBIS EDITORIAL TASK GROUP http://www.ibis.org/Editorial_wip/ Mailing list: ibis-editorial@freelists.org Archives at http://www.freelists.org/archive/ibis-editorial/ ====================================================================== Attendees from March 24 Meeting (* means attended at least using audio) ANSYS Curtis Clark Intel Corp. Michael Mirmak* Keysight Technologies Radek Biernacki*, Ming Yan Mentor Graphics Arpad Muranyi* Micron Technology Justin Butterfield, Randy Wolff* Signal Integrity Software Walter Katz*, Mike LaBonte* Teraspeed Labs Bob Ross* Michael Mirmak convened the meeting. No patents were declared. Mike LaBonte moved to approve the minutes of the March 17 meeting. Bob Ross seconded the motion. The minutes were approved without objections. During the AR review, Michael noted that he has yet to send the curve-tracing measurement drawings to the reflector. Mike & Bob continue to work together to resolve language in BIRD181 and referencing text; this is addressed in a presentation to be discussed later in the meeting. Walter Katz has fulfilled his AR to send out his own proposals for new language on referencing. Mike continues work to adjust the text surrounding his new IBIS 6.2 figures, separately from his BIRD181.1 work. Mike added that the requirement section of the BIRD is done; the document still needs adjustments to the figures and text, which can be reviewed in the next meeting. Mike checked on the submission date for BIRD 181.1 (currently listed as Oct. 2016). The BIRD on-line has dates for 181, which were off by a few days each. All dates on-line are correct now. Work on a draft BIRD 181.2 is in progress. Bob presented material summarizing issues with BIRD 181.1. He suggests that the nature of the changes is such that the Open Forum could delay the changes until later and leave IBIS as-written. He also discussed several issues aside from BIRD181.x that need resolution. On referencing, terminals are used in many cases for referencing that are usually intended to be at the buffer. For I-V tables, the equations in the original specification were only to intended to describe polarity conversion of the table data, not referencing per se. Michael asked whether the new full equation approach is less intuitive. Bob suggested that it was. Radek Biernacki agreed with Bob's proposed changes in "Other Notes" portion of IBIS, adding that current needs to be defined there. Bob noted that current conventions defined elsewhere in the document need to be re-checked. Bob also noted that the typ/min/max combinations in the equation need to be taken into account. Simiarly, typ/min/max varitions are needed in [ISSO P*] figures and defintions. There is a discrepancy regarding the figure (which shows typ/min/max) and the text (which suggests typical-only). The team should consult older documentation and presentations on the ISSO keywords to see original definitons. Bob added that in the I-V section, there are too many references in the table Range definitions to CMOS as opposed to other technolgies. Bob suggested that moving the Notes on Data Derivation material to the I-V section is workable, but may have other impacts. He added that the Table of I-V Table Ranges for Low and High Voltages needs clarification. Radek noted that the table is in the context of the pullup and pulldown, not the table reference. Bob concluded by stating that all of these changes are required in order to adopt nodal terminology consistently throughout IBIS. Michael asked whether a nodal approach is what we want to do; IBIS does not necessarily require use of SPICE syntax. Walter stated that IBIS should indeed be consistent with SPICE syntax. He suggested avoiding writing a BIRD, as a complete document with edits can be generated and submitted more easily, but will defer. Bob advocated writing at least one BIRD to address the issues clearly. Walter suggested that Bob write the BIRDs. Michael asked whether the number of BIRDs is tractable. Bob replied that at least 4 more BIRDs would be needed, complicating decisions about IBIS versions. Bob added that another option would be to approve BIRD 181 as-is, then leave the rest of IBIS as-is, to be edited later. Mike summarized the recent history, observing that BIRD181 fixed table issues, but then nodes, voltage values, etc. were noted to be inconsistent in their treatment elsewhere. Then Bob realized that Notes on Data Derivation is also inconsistent and needs to be fixed. Total consistency would require a very large update. According to Mike, Walter had suggested that IBIS 6.2 was intended to do this. Mike suggested that the I-V table section and Notes on Data Derivation section remain in BIRD181. Mike's figures BIRD would be completely separate and is intended to show full circuits with clean terminology. Mike agreed with Bob that we need to "show our work" in a BIRD or BIRDs that document the rationale for any changes. Walter moved to adjourn. Bob seconded the motion. The meeting adjourned without objection.