====================================================================== IBIS FUTURES/COOKBOOK SUBCOMMITTEE MEETING MINUTES Date: April 22, 2004 Attendees: Cadence Design Systems - Lance Wang Cisco - Syed Huq Green Streak Programs - Lynne Green Intel Corp. - Michael Mirmak, Arpad Muranyi Mentor Graphics - John Angulo, Ian Dodd Teraspeed Consulting - Bob Ross ====================================================================== Next Meeting: Thursday, May 6 1 PM - 3 PM US Pacific Time Telephone Bridge Passcode 916-356-4261 4 282-0276 Agenda: Cookbook Discussion/Opens Futures Discussion/Opens - Continuation of User-Definable Measurements (Muranyi/Wang) The 2-3 PM hour is reserved for parser bid discussions. ====================================================================== The committee meetings began with a brief discussion regarding arrangements for DAC. COOKBOOK MINUTES Briefly discussed were proposals from Lynne to combine EBD, PKG and ICM guidance into one Cookbook document. The participants agreed with the caveat that this document be separate from the IBIS Cookbook. No other Cookbook agenda items were discussed. ARs: John Angulo - provide overview of differential model extraction details Michael Mirmak - create outline using current examples of above model types ====================================================================== FUTURES MINUTES Arpad and Lance reviewed their proposals regarding a user-definable measurement and delay specification. While technically possible, including [Model]-style measurement parameters under [External Circuit] is not a realistic option for future spec development. Instead, a SPICE-like format was proposed, to include the domain (AC, Time-Domain, etc.) of the measurement and equations for probing at nodes. Lance further suggested that such a measurement scheme include constraints and even custom reporting formats. Ian Dodd suggested that this was perhaps too SPICE-like and that the purpose of user-definable measurements would be to abstract away SPICE nodal measurement descriptions. Lynne Green further commented that much of what was being described was post-processing rather than true measurements. Additionally, encrypted internal blocks may prevent measurement in many cases. Bob Ross suggested that a design kit approach may be the best way of providing measurement equation information, on a per-design basis. Additional issues raised included whether multiple buffers could be referenced within CIRCUITCALL and whether measurements should be conducted at a component level or at the buffer or pad level. Michael summarized his IIRD1 proposal, to ensure clear usage of S-parameter data by ICM-compatible tools. Participants agreed that MULT=1 should be required and issuing an error would be appropriate Syed summarized his concerns regarding I-t curves and power integrity. At present, Cisco is interested in seeing a buffer-level change to the IBIS specification to include current distribution data, to be extended to the core level in the future. Istvan Novak was hoping that a standards body such as IBIS could lead the way in proposing a standard format. Arpad Muranyi suggested that series elements, already in the specification, might enable current characterization, if permitted between power and ground rails. Alternately, a noise source element could be added to the specification. Bob Ross suggested that vector file definitions might be required to fully characterize power vs. core and buffer events. Discussions will continue in upcoming meetings. An open issue is whether BIRD42.3 should be re-opened for consideration. ARs: Ian - what do current applications need for measurements (ex. serial differential, etc.) Arpad - same question, Intel perspective ---------------------------------------------------------------------- For reference, the seven original BIRD proposals are listed below: 1) Define accepted digital states for input and output ports on IBIS 4.1 AMS models 2) Provide explicit links between ICM and IBIS 3) Expand [Circuit Call] to accept native IBIS [Model] and [External Model] references 4) Define series links between pins, pads and/or nodes under the multi-lingual extensions 5) Create user-definable buffer delay and input threshold measurements 6) Permit re-use of port names under [External Model] 7) Permit native IBIS data to be linked through external file names, similar to the multi-lingual extensions Attendees were assigned to complete formal proposals or definitions for the above, as follows: 1) Ian Dodd 2) Michael Mirmak 3) John Angulo 4) Ian Dodd & Bob Ross 5) Lance Wang, Ian Dodd and Arpad Muranyi 6) Arpad Muranyi 7) Lynne Green The formal statements should include a problem statement and a general outline of the solution. I-t and core currents will be discussed by Syed at a future meeting.