8/26/94 IBIS Open Forum Meeting Minutes

Text item: Text_1

Date: Sep. 6, 1994

From: Will Hobbs (503) 696-4369, fax (503) 696-4210
         Will_Hobbs@ccm2.jf.intel.com
         XTG Modeling Manager, Intel Corp., Chairperson, IBIS Open Forum
         Intel Corporation
         5200 NE Elam Young Pkwy, Hillsboro, Oregon 97124 USA
         and
         Derrick Duehren (503) 696-4299, fax (503) 696-4904
         Derrick_Duehren@ccm.jf.intel.com
         Intel Program Manager

Subject: Minutes from IBIS Open Forum Meeting 8/26/94

To:
AT&T Global Info Systems Dave Moxley*
Anacad Steffen Rochel
Ansoft Henri Maramis
Atmel Corporation Dan Terry
Cadence Design Sandeep Khanna, C. Kumar*
Cadlab Ralf Bruning
Contec Maah Sango
Digital Equipment Corp. Barry Katz
EIA Patty Rusher
High Design Technology Michael Smith, Dr. Ing. Cosso
HyperLynx Kellee Crisafulli
IBM Jay Diepenbrock, Joseph Flanigan
IBM-Motorola alliance Lynn Warriner, Hoa Quoc, John Burnett
Integrated Silicon Systems Eric Bracken
Integrity Engineering Greg Doyle
Intel Corporation Stephen Peters*, Don Telian, Will Hobbs*
                              Arpad Muranyi*, Derrick Duehren*,
Interconnectix, Inc. Bob Ross*, Scott Bloom*
Intergraph Ian Dodd, David Wiens
IntuSoft Charles Hymowitz
Mentor Graphics Ravender Goyal
Meta-Software Mei Wong, You-Pang Wei, John Sliney
NEC Hiroshi Matsumoto
MicroSim Arthur Wong
National Semiconductor Syed Huq*
North Carolina State U. Steve Lipa, Michael Steer
OptEM Engineering, Inc. Benny Leveille, Ken Ehn
PC Ware Paul Munsey
Quad Design Jon Powell*
Quantic Labs Mike Ventham
Racal-Redac John Berrie
Siemens Nixdorf Werner Rissiek*, Olaf Rethmeier
Symmetry Martin Walker
Synopsys, Logic Modeling G. Randy Harr
Texas Instruments Bob Ward*
Thomson-CSF/SCTF Jean Lebrun
UniCAD Canada Ltd. Stephen Lum
Zeelan Technology George Opsahl, Hiro Moriyasu

CC:
Intel Corporation Randy Wilhelm, Jerry Budelman,
                              Intel IBIS team

In the list above, attendees at the meeting are indicated by *.

Upcoming Meetings: The room and bridge numbers for future IBIS teleconferences
are listed below:
     Date Bridge Number Reservation #
     9/16/94 (415) 904-8944 #809164
     10/7/94 (415) 904-8944 #809164

All meetings are 8:00 AM to 10:00 AM PDT (15:00 to 17:00 UTC). 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.

NOTE: "AR" = Action Required.

------------------------------------------------------------------------

Check-in, Intros, Announcements
Bill denBeste introduced himself and reported that he has taken over as
administrator for the vhdl.org machine (IBIS and DIE), as a contractor in
place of Randy Harr, who has accepted a position at Stanford U. Bill can be
reached at (503) 649-7477, billd@reprise.com

There were no corrections made to last month's minutes.

Will paid Paul Munsey for his evaluation of the 2.0 changes ($250).

There were no new agenda items.

There is an IBIS mention in the current ASIC and EDA System Design, Pg. 34.

Progress toward enlisting new IC vendors
There was no new public model availability announced. Jon is working with
VLSI, they have a program in place to create IBIS models. Will reported that
one of Intel's customers reported that a DRAM supplier provided IBIS models
with their components. This supplier is not one of the regular forum members.
Bob Ward reported that he has nearly succeeded in working through TI politics,
IBIS consensus is converging. Syed reported that he presented IBIS to
National Semiconductor. They are now moving toward incorporation of IBIS into
their release process.

AR Will: Track down the DRAM IBIS vendor, see if they will post the models.

Golden Parser, 2.1
Paul was not present in the meeting, but sent the following update:

  Ron Neville and I are moving forward. We have most of the data
  structure figured out, but are still ironing out a few wrinkles. I, for the
  most part, just need to package it all together into one file so I can mail
  it out for those who are interested in looking at it. At the same time we
  are also progressing with the coding in the areas that we have well defined.

  It is proving very beneficial to have 2 people working on this project,
  sharing ideas, and coming to an understanding of the best methods and design
  for the parser. I do think we're behind where we would like to be at this
  time, but we are still targeting the end of September. The internal design
  is changing some from V1.1 to accommodate the requirements and better fit
  the V2.0 and future enhancements. - Paul Munsey

Timing the release of 2.1 models
Kumar, Jon, and Bob Ross confirmed that they, as simulator vendors, will be
ready to support 2.1 models when the 2.1 Golden Parser is ready (November).
To do so, however, they will need 2.1 models earlier. They would like early
copies from model suppliers.

Jon proposed that IC vendors post perturbed 2.1 models as examples to vhdl.org
/models directory by the end of September. Intel and National committed to
provide the examples; TI thinks they can do so, also.

We recognized/acknowledged that 2.1 models can contain lots of data that can
be used in many creative ways.

The most important version 2.1 features to semiconductor vendors follow:

Intel: multiple rails (various voltage rails), non-linear ramp tables
National: multiple rails, differential I/O
TI: multiple rails, differential I/O, multiple and non-linear ramp tables.

IBIS Librarian
Michael Steer of NCSU can't make Friday meetings during the academic year. He
had been unavailable for a few weeks, but is back now.

Old AR Michael: Post procedures for companies to post IBIS files.

Old AR Michael: Propose a longer, more thorough disclaimer statement and send
it to Derrick Duehren for legal review.

Standards Body Affiliation
Will summarized previous discussions: affiliation is good, it legitimizes IBIS
(IBM comments), provides legal standing and international exposure.

Bob Ward and Jon Powell added their votes for affiliation with EIA.

AR Will: Post notice to the reflector of our intent to affiliate so if anyone
has a problem with it they can react.

Bill denBeste recommended having an IBIS participant in the DIE meetings. DIE
wants to go out for vote on the spec. in October. We would like to tag along,
but we will not be ready until probably November. Bill reported that the DIE
meeting on 9/13, 9/14 will cover schedule and EIA affiliation. DIE ARPA
funding will conclude in Dec. or Jan.

AR Bill: Post DIE 9/13/14 meeting agenda to the reflector. [DONE]

Spice-to-IBIS Converter
Bob Ross reported that the V1.1 converter is available on vhdl.org. It works
very well if the Spice models work under IBIS ranges. There are a few
undocumented glitches. Some Spice models aren't built to sweep over the IBIS
range, causing glitches. S2ibis works well if the Spice model is built to
handle the extreme ranges. When the Spice run doesn't converge, S2ibis tells
the user when and where the non-convergence occurred. There is a man page, a
spec., and a readme file with the program, but they don't cover the gotchas.
There is also a try.me file to use with the S2ibis program. Will suggested a
public comments file for people to contribute to. The forum decided to post
experiences to the reflector and copy Michael Steer at NCSU
(mbs@eos.ncsu.edu).

AR All: Try out the converter and post comments to the reflector. If nothing
else, please send email to Michael. He can use it when seeking additional
ARPA funding.

2.0 BNF
Bob Ross would appreciate any comments on the BNF. Bob Ward will post a
response as soon as he has reviewed it. Bob Ross believes that we should have
a very robust 2.0 BNF available.

1995 Design Automation Conference
Kumar talked to Paul about DAC'95. We should submit as many papers as
possible. It should describe the design process and how IBIS is used. Kumar
and Werner are planning to submit papers. We may need to submit 20 papers to
have 5 accepted to qualify for a slot.

AR Jon: Start a discussion on the reflector on this subject. [DONE. We only
need 3 or 4 well-written papers]

BIRD 17, Number of Data Points
The hundred-point limit debate continues. Arpad put the 100 point limit in
originally somewhat arbitrarily. Will is inclined to kill the BIRD, Jon
wants no limit due to customer input, though he thinks 50 is enough. Arpad
feels that 151 would be a better choice than 101 to cover -5 to +10 in .1 volt
increments. Consensus was reached to kill the BIRD.

Vote taken: BIRD 17 unanimously NOT approved.

BIRD 19, V_fixture sub-keyword
Concern over rising and falling waveform timing with a shifting power supply.
Bob proposes this BIRD for inclusion in the 2.1 IBIS spec.

Vote taken: BIRD 19.1 unanimously approved for inclusion in 2.1.

Voltage range "span"
"Span" in the spec. is inclusive and allows voltages to go beyond the range.
Arpad can send a test case to Jon if he would like his code tested.

Wrap-up, Next Meeting Plans
Our next meeting is 9/16/94.

==============================================================================
                                      NOTES
If you know of someone new who wants to join the e-mail reflector
(ibis@vhdl.org), send e-mail to ibis-request@vhdl.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 (408-945-4170), or send an email request to the automatic
archive server, archive@vhdl.org.
==============================================================================
Received on Tue Sep 6 11:48:26 1994