DATE: October 23, 1996 SUBJECT: 10/18/96 EIA IBIS Open Forum Minutes VOTING MEMBERS AND 1996 PARTICIPANTS LIST: AMP Hank Herrmann*, Tim Minnick, Russ Moser, Ray Ziesse*, Jeff Walden* Applied Simulation Technology Dileep Divekar*, Norio Matsui, Antonis Orphanou (formerly Contec) Cadence Design C. Kumar* Cypress Bruce Wenniger Hewlett Packard, EEsof Karl Kachigan, Henry Wu HyperLynx Kellee Crisafulli INCASES Olaf Rethmeier, Ralf Bruening Intel Corporation Stephen Peters*, Will Hobbs, Arpad Muranyi*, John Keifer, Aaron Tang, Duane Quiet, [Donald Telian], Jim Kruchowski Interconnectix, Inc. Bob Ross*, Chris Reid Meta-Software (Sanjay Gangal) Mitsubushi Tam Cao, Hoang Nguyen Motorola Ahmed Omer National Semiconductor Syed Huq*, [Donald Snyder], Chune-Sin Yeh, [Bill Aronson] NCR (formerly ATT-GIS) Dave Moxley*, Richard Mellitz NEC (Hiroshi Matsumoto), Quad Design Jon Powell, Chris Rokusek* Quantic Laboratories (Mike Ventham) Texas Instruments Roger Cline, Ben Andresen, Sri Jandhyala, Tareq Shahwan, Dan Phipps Thomson-CSF/SCTF (Jean LeBrun) UniCAD Canada Ltd. Celso Faia Veribest Ian Dodd*, David Wiens VLSI Technology Dick Ulmer, Sung Oh, Swami Gangadharan, Daniel Kim, Tom Dockery, D.C. Sessions, Hrish Patel Zuken-Redac (John Berrie) OTHER PARTICIPANTS IN 1996: 3M Fran Hart*, George Hare* Alcatel John Fitzpatrick Altera Vadim von Brzeski Apteq Design System Dan FitzPatrick Compaq Mark Leonard Digital Equipment Corp. Jeff Chu* EIA Patti Rusher IC Works Eric Chen Mentor Graphics Kim Owen Micron Technology Brian Johnson MicroSim Ralph Perez Molex Gus Panella North Carolina State U. (Michael Steer) Oki Semiconductor Tom Chao Philips Semiconductor Mike Magdaluyo Rockwell Collins Thomas Thurman S3, Inc. Sarathy Sribhashyam, Phap Duong Seimans Nixdorf Arnaud Lacariere Symmetry Andy Hughes Tektronix, Inc. Brock Hannibal TRW & Free Model Foundation Robert Harrison VTC, Inc. Bob Ward Zeelan Technology George Opsahl In the list above, attendees at the meeting are indicated by *. Principle members or other active members who have not attended are in parentheses. Participants who have joined another organization are in square brackets. Upcoming Meetings: The bridge numbers for future IBIS teleconferences are as follows: Date Bridge Number Reservation # Passcode 11/08/96 (916) 356-9200 2-104708 1692479 All meetings are 8:00 AM to 9:55 AM Pacific Time. We try to have agendas out 7 days before each open forum and meeting minutes out within 7 days after. When you call into the meeting, ask for the IBIS Open Forum hosted by Will Hobbs and give the reservation number and passcode. NOTE: "AR" = Action Required. -------------------------------- MINUTES ------------------------------------- INTRODUCTIONS Jeff Walden from AMP works with Ray Ziesse and is involved with connector modeling. George Hare from 3M is in the Electrical Products Division in Texas and has been working on modeling issues for the last six months. EIA MEMBERSHIP AND TREASURER'S REPORT Bob Ross reported that the EIA IBIS ledger remains at $9,285 for September, 1996. However some EIA shared expenses will be deducted by the end of the year. MINUTES REPORT, MISC. Fran Hart reported that she is with the Electronic Products Division of 3M, not the Electronics Products Group as previously stated. MISCELLANY/ANNOUNCEMENTS Bob Ross reported that each IBIS reflector mailing list contains over 290 people from all over the world. Each mailing typically generates a few bounced messages, typically because of people moving on and because of local computer network problems. Bob usually tries to forward the mail in a few days or attempts to contact the site "postmaster". However, if the problem is not resolved in a few days, the name is removed on the list. So, if you find that you are not receiving messages, you may have been dropped. One way to find out is to check the e-mail archives (email.archive and monthly archives) to see if you have received the most recent messages. PRESS AND WEB PAGE UPDATES Syed Huq reported that within a few weeks he will add links to an HTML version of the EE Times Article and to the BIRDs and the Cookbook on vhdl.org. He also plans some format changes on the official EIA/IBIS home page. Bob Ross also reported "PCB design tools continue Windows juggernaut", Computer Design, September 1996, pp. 35-38 contain reference to VeriBest usage of IBIS models. NEW MODELS Arpad Muranyi has several more Intel models available under NDA. He will submit the updated list to Jon Powell. Bob Ross reported that the Free Model Foundation now has 29 contributed IBIS models under ftp://vhdl.org/vi/fmf/fmf_public_models/ibis Arpad also raised the issue that the response time on vhdl.org is very slow. Bob has experienced reasonably fast responses, so he feels that the problem may possibly be internal to Intel. In any case both will try again to see if there is a response problem. [Bob and Arpad both found the response normal.] OPENS FOR NEW ISSUES Ray Ziesse on AMP patent position. C. Kumar on V/T waveform problems with IBIS Version 2.1. AMP PATENT POSITION Ray Ziesse responded to an earlier commitment that AMP provide its position to the Committee concerning Patent # 5,081,602 on Computer Simulator for Electrical Connectors. (Refer to the September 6, 1996 IBIS Open Forum Minutes.) A licensing agreement is available on Patent # 5,081,602. It can be obtained by contacting. Mr. Bruce Wolstoncroft The Whitaker Corporation Phone: (302) 633-2745 DESIGN SUPERCON97 Syed Huq reported that National Semiconductor has formally offered to host the IBIS Face-to-face meeting in Santa Clara on the Monday before Design SuperCon97 at the end of January. National will provide the room, lunch and support equipment. Syed felt that an EIA procedural rule prohibited a company such as National from hosting a formal EIA meeting on company premises. [This is not true, and in fact is encourage, based on a response from Patti Rusher.] We expect to formally select the host company at the next Open Forum meeting. Anyone else wishes to be considered should notify Bob Ross. Bob expects the meeting agenda to include some practical experiences regarding generating IBIS models by silicon vendors and by customer's with internal ASIC development capability. EDA vendors may be able to share more information on internal utilities. Bob also expects more discussions on model validation. Arpad Muranyi followed up on a point made at the last meeting that the Committee should consider holding some meetings on the East Coast for geographical balance. Bob indicated that the two traditional Face-to-face meetings are held at the Design SuperCon location (Santa Clara, California) and at wherever the Design Automation Conference (DAC) is held. The location in June 1997 is in Anaheim, CA). Many active IBIS Committee members already attend these conferences. So the opportunity for meetings elsewhere depend on DAC locations. Possibly some working group Face-to-face meetings can be held, if needed, on the East coast. DAC97 PLANNING (AND OTHER MEETINGS) Bob Ross indicated that EIA usually plans to have a Standards Booth covering the activities of all committees under EIA/EIG. The IBIS Open Forum Committee shared expenses with other standards groups last year, and we will probably do so again this year. S2IBIS PLANS Based on recent reflector activity, there are several questions and problems with s2ibis2. Bob Ross reported that the original s2ibis Version 2.091 was posted as a BETA version on vhdl.org around November, 1995. Some significant concerns including crashes, functional errors, and functional omissions have been reported. NCSU has used all their funding for the existing development and has not gotten new funding from DARPA to finish the job. NCSU has also investigated commercial opportunities. Bob offered several options to get the s2ibis2 code beyond the prototype level and to be functionally usable: 1. NCSU "steal" some time to complete the job. 2. Fund a new student for several months to complete the job. 3. Some other company take over "ownership" and complete s2ibis2. 4. Some individual be contracted to finish the job. Ian Dodd indicated that VeriBest has made its own corrections and is willing to communicate them publicly. VeriBest may offer their s2ibis conversion utility to the public domain. They will report some significant problems that they have discovered on the IBIS reflector. Syed Huq felt that NCSU should be encouraged to finish the job. Bob felt that while there are several items that need to be fixed, the completion project was still of limited scope. There is interest in an NT based utility. However, Bob felt this would be beyond the scope of this project. AR - Bob Ross work with NCSU to correct some defects and complete s2ibis2. AR - Ian Dodd post the problems he discovered. Optionally, Ian work with Bob Ross to post a working revised s2ibis2 conversion package if he obtains permission. Arpad Muranyi mentioned that his IBIS Center development utility is available to IBIS Committee members. He is restricted by a licensing agreement to 50 copies. Arpad cannot provide support but would welcome comments and feedback. The utility is not general. It focuses on CMOS technology and does not, for example, support ECL technologies. It does the following: 1. Reads simulated data from template files 2. Graphically displays the simulated data 3. Creates an IBIS Model set of data 4. Graphically displays the IBIS Model tables 5. Construct an IBIS Component model from several Models The graphical interface supports zooming. Templates are provided to set up the simulations to generate the initial simulated data, as a separate process. The utility is available for Sun Unix and DOS workstations. IBIS V/T PROBLEM WITH VERSION 2.1 IBIS MODELS C. Kumar raised the issue that some Version 2.1 level IBIS models contain V/T data that converge slowly to the final DC value. One issue is what may be a reasonable truncation criteria so that one can properly focus on the faster transition portion. Several people suggested experience using one percent. Kumar questioned whether this should be standardized. Too much truncation creates a DC mismatch. Chris Rokusek stated that the percentage voltage versus time approach can produce stable results. Too much truncation may hide other artifacts related to noise and EMI. Any criteria could be put in a cookbook for guidance. How C_comp interacts is a related issue. An actual fixed C_comp used in the IBIS model differs from any non-linear capacitance in the device or Spice model. These capacitances cannot be removed. Therefore, it is realistic that a simulation could differ from a "golden response". The point about the golden response is that it is a response based on a prescribed condition. Simulators probably have different algorithms to describe transitions, so one test of accuracy is how close the response is to the "golden response". AR - Kumar write up the specific issues and concerns on the IBIS reflector. BIRD37.2 Stephen Peters presented BIRD37.2 as a modification to BIRD28.3 which extended the package model by allowing cascaded sections and cascaded matrix sections. The matrix addition is removed. A simple Fork and End Fork branching capability is added. Per a Package Committee request, Bob Ross gave background of the existing package issues under discussion. Package models extensions consist of BIRD28.3 and BIRD37.2. At the other extreme electrical boards can be described by proprietary physical data bases and is being standardized by EDIF 4 0 0. The IBIS committee has been interested in board electrical descriptions for such components as SIMMs and MCMs. These descriptions would not be for large boards. They would allow models where the physical data is is not available (and for boards being designed). Transmission line models are appropriate. Several formats have been proposed: nodal descriptions (Spice-like), transmission line connection syntax (Z0, TD, Length, Forks), and field extractor like (L, C, length) formats. In practice, single line structures are quite adequate for electrical boards for timing and signal integrity analysis. Connector models are another requirement. Single line models can be used. However there is strong interest in multi-line (coupled) models to support multiboard applications. The typical analysis may cover cross-talk, ground bounce and differential nets. BIRD36 covers both the single line electrical board descriptions and also multi-line connector models. BIRD31.3 is a connector only proposal that is an extension of BIRD28.3 to cover two-terminal devices. It is currently on hold pending the resolution of BIRD36. BIRD32 is a coupling proposal extension that deals with subcircuit matrices as a more economical way of formatting matrix coupling. This is also on hold. Thus the proposals currently under consideration are BIRD37.2 for packages and BIRD36 for electrical board descriptions including connectors. Several minor editorial corrections were suggested for BIRD37.2. BIRD37.2 was approved unanimously. AR - Stephen Peters issue approved BIRD37.3 with the corrections. [Done] PACKAGE COMMITTEE REPORT Stephen Peters reported on Package Committee discussions. One consensus is that the electrical description is positioned to support smaller boards is not intended to be as comprehensive as a full physical description. So uncoupled networks are adequate. The EDIF 4 0 0 format covers the full physical description including coupling and contains references to IBIS models. Connectors are typically a repetitive structure that can be described by simple, single line, pin-to-pin formats and also by coupling matrix formats. EDA vendors need coupling. So the question is how to describe the coupling. Hank Herrmann has made some changes to BIRD36 based on some previous meeting comments and suggestions. He will forward the revised BIRD36 to Stephen to distribute to the Package Committee. Hank noted that single line connector models could be provided immediately once the format is defined. Package Committee meetings will continue on Fridays, 8:30 A.M. to 10 A.M. AR - Stephen Peters set up the Package Committee meeting. SPECIFICATION COMMITTEE REPORT Bob Ross reported that Ahmed Omer, John Fitzpatrick, Jon Powell and Bob held several conference call meetings resulting in BIRD39. Its key aspects are: - A new optional specification keyword [Model Spec] under [Model] - [Model Spec] subparameters with typ, min and max columns - Vinl and Vinh thresholds extended to min and max - Hysteresis thresholds - Overshoot and dynamic overshoot parameters - Pulse immunity parameters Arpad Muranyi questioned whether an area or energy definition could be used for dynamic overshoot. Bob responded that the committee had considered such an approach. The problem was in obtaining a value. Also the committee found that the source data information came in all types of formats. For example a PCI bus overshoot specification is derivable from a current limit value and clamping diode details. Some of the information may be in general sections showing typical characteristics rather than associated with a specific part. So the Committee opted for a simpler method where a voltage can exceed the overshoot value for a specified period of time as long as it remains below the dynamic overshoot limit value. The pulse immunity values also provide a do-not-exceed value for a specified period of time beyond some existing thresholds. A vote will be scheduled at the next meeting. Also a vote to reject BIRD38 will be scheduled since BIRD39 supersedes BIRD38. BIRD35.2 - MULTI-STAGED OUTPUTS Bob Ross recently introduced BIRD35.2 proposing a revised scheduling method He believes it is compatible to existing implementations, but plans to check with Jon Powell. EGG10 - PARSER ADDITIONS FOR NUMERICAL CHECKING Chris Rokusek indicated he could implement the proposed changes in one module so that the revised code could provide a new basis for further development. The unchanged ibischk2 would still be available. Chris and Bob Ross need to decide on some functionality details mentioned at the last meeting. AR - Chris Rokusek and Bob Ross decide on some functionality details and Chris produce the revised ibischk2. NEXT MEETING: The next meeting is on Friday, November 8, 1996, 8:00 A.M. to 9:55 A.M. BIRD39 and BIRD38 are scheduled for a vote. ============================================================================== NOTES IBIS CHAIR: Bob Ross (503) 603-2523, Fax (503) 639-3469 bob@icx.com Modeling Engineer, Interconnectix, Inc. 10220 SW Nimbus Ave, K4, Portland, OR 97223 VICE CHAIR: Syed Huq (408) 721-4874, Fax: (408) 721-4785 huq@rockie.nsc.com Staff Applications Engineer, National Semiconductor, M/S A-2595 2900 Semiconductor Drive, Santa Clara, CA 95052 LIBRARIAN: Jon Powell (805) 988-8250, Fax: (805) 988-8259 jonp@qdt.com Transmission-Line Products Manager, Quad Design 1385 Del Norte Rd., Camarillo, CA 93010 SECRETARY: Vacant This meeting was conducted in accordance with the EIA Legal Guides and EIA Manual of Organization and Procedure. The following e-mail addresses are used: ibis-request@vhdl.org To join, change, or drop from either the IBIS Open Forum Reflector (ibis@vhdl.org), the IBIS Users' Group Reflector (ibis-users@vhdl.org) or both. State your request. ibis-info@vhdl.org To obtain general information about IBIS, to ask specific questions for individual response, and to inquire about joining the EIA-IBIS Open Forum as a full Member. ibis@vhdl.org To send a message to the general IBIS Open Forum Reflector. This is used mostly for IBIS Standardization business and future IBIS technical enhancements. ibis-users@vhdl.org To send a message to the IBIS Users' Group Reflector. This is used mostly for IBIS clarification, current modeling issues, and general user concerns. ibischk-bug@vhdl.org To report ibischk2 parser bugs. The Bug Report Form Resides on vhdl.org in /pub/ibis/bugs/bugform.txt along with reported bugs. Information on IBIS technical contents, IBIS participants, and actual IBIS models are available on the IBIS Home page found by selecting the Electronic Information Group under: http://www.eia.org Check the pub/ibis directory on vhdl.org for more information on previous discussions and results. You can get on via ftp anonymous, "guest" login from telnet or dial-in (415-335-0110), or send an e-mail request to the automatic archive server, archive@vhdl.org. "IBIS Spoken Here" placards are available from Jon Powell (jonp@qdt.com) for use at trade shows. ==============================================================================