====================================================================== IBIS FUTURES/COOKBOOK SUBCOMMITTEE MEETING MINUTES Date: March 10, 2005 Previous Meeting Attendees: ---------- Cadence - Donad Telian, Lance Wang Cisco - Vinu Arumugham, Syed Huq, Zhiping Yang Intel - Michael Mirmak, Arpad Muranyi Mentor - John Angulo Samtec - Otto Bennig Teraspeed - Bob Ross ====================================================================== 1 PM - 2 PM US Pacific Time Telephone Bridge Passcode 916-356-2663 2 931-5695 Agenda: 1 - 1:05 PM Opens 1:05 - 1:30 PM ICM Parser Closure IBIS 4.1 Parser Closure BIRD94, 97 Discussion (note that Arpad cannot attend this meeting) 1:30 - 2:15 PM BIRD95.2 Update (Syed Huq) 2:15 - 3:00 PM ICM-IBIS Links Update ====================================================================== Futures Minutes (abbreviated discussion) Team discussed Bob Ross's comments on BIRD95.1. Most comments could be addressed offline. Bob mentioned, in his note 13, that total buffer current can be extracted from the ground side supply and converted to power rail current. "Rail" composite current may be hard to separate, however, according to Zhiping. Arpad noted that he was working on a gate modulation effect BIRD, which should be ready before the next summit. Donald summarized his material from the summit and Open Forum on SPICE macromodeling as a new method of behavioral modeling. Michael presented a set of questions and concerns on SPICE macromodeling and behavioral modeling's future. Bob Ross inquired, to Donald, about SPICE linkages and the equations embedded within SPICE tools; these equations would not be contained in the model itself, as would be the case with AMS. Berkeley SPICE also has an existing equation format; how will this be supported? Donald noted that this format is not substantially different than HSPICE. Arpad and Bob observed that HSPICE tends to have more features. Donald also noted that the interpretation of Berkeley SPICE support held by the industry might be different than that intended by the Open Forum: the specification might have intended to support only passive interconnects (on-die and package), but the industry interpreted this as a move toward transistor and macromodel support. Donald doesn't have trouble with implicitly supporting SPICE-to-SPICE conversion (i.e., converting from one SPICE format to another, possibly proprietary, format); all vendors are experienced with it. No Cookbook meeting was held.