EIA IBIS Summit Meeting Minutes

Date: 2/2/01

SUBJECT: 1/29/01 EIA IBIS Summit Meeting Minutes

VOTING MEMBERS AND 2001 PARTICIPANTS LIST:
3Com (& CommWorks) Roy Leventhal
Agilent (Mark Chang)
Ansoft Corporation (Eric Bracken)
Applied Simulation Technology Raj Raghuram*, Norio Matsui*, Fred Balistreri*
Avanti (Chen Hongyu)
Brocade Communications Robert Badal*
Cadence Design Ian Dodd*
Cisco Systems Syed Huq*, Lungfu Chen*
Compaq Peter LaFlamme*, Ron Bellomio*, Quang Dam*
Cypress (Rajesh Manapat)
EMC Corporation Brian Arsenault*, Jinhua Chen*
Fairchild Semiconductor Adam Tambone*
IBM Michael Cohen*, Greg Edlund, Wes Martin*,
                               Yeon-Chang Hahm*, Bill DeVey*
Innoveda (& HyperLynx) Guy de Burgh*, John Angulo*
Intel Corporation Stephen Peters*, Arpad Muranyi*, Dave Lorang,
                               Michael Mirmak, Qinglun Chen*, Will Hobbs*
LSI Logic Larry Barnes*
Mentor Graphics Bob Ross*, Tom Dagostino*, Chris Reid*,
                               Mike Donnelly*, Hazem Hegazy*, Tony Dunbar*,
                               Griff Derryberry*, Dan Lake*, Sherif Hammad*,
                               Mohammed Korany*, Weston Beal*
Micron Technology Randy Wolff, Yong Phan
Mitsubishi (Tam (Tom) Cao)
Molex Incorporated Gus Panella*, Brian O'Malley*
Motorola (Ron Werner)
National Semiconductor Milt Schwartz*
Nortel Networks Calvin Trowell
North East Systems Associates Edward Sayre*
Philips Semiconductor Zack Ciccone*
Quantic EMC (Mike Ventham)
Robinson-Nugent, Inc. (Alexander Barr)
Siemens AG Bernhard Unger*, Helmut Katzier*
SiQual Scott McMorrow*, Rob Hinz*, Bernard Voss*,
                               Chris Brewster*
Texas Instruments Thomas Fisher*, Stephen Nolan*, Ramzi Ammar*
Time Domain Analysis Systems Dima Smolyansky*, Steve Corey*
Tyco Electronics (Russell Moser)
Via Technologies (Weber Chuang)
Zuken (& Incases) (Werner Rissiek)

OTHER PARTICIPANTS IN 2001:
Actel Corporation Silvia Montoya*
Acuson Kim Helliwell*
AMCC Jeff Smith*
Apple Computer John Figueroa*
ASIS Ltd David Wright*
EIA Cecilia Fleming
FCI Sercu Stefaan*
Foundary Networks Bertram Chan
Framatom Conectors Danny Morlion*
Fujitsu Ltd Tadashi Arai*, Takeshi Murakami*
Huawei Technologies Rachild Chen*
Hyundai Electronics Jongho Kang*
Intrinsix Corporation Steven Chin*
Nokia Tapani von Ravner*
Oak Technology Darmin Jin*
Plexus Technology Group Joseph Socha*
Signal Integrity Software Douglas Burns*, Barry Katz*, Walter Katz*
STMicroelectronics Peter Hirt*
Xilinx Susan Wu*
Independent, Consultant Al Davis*

In the list above, attendees at the meeting are indicated by *. Principal
members or other active members who have not attended are in parentheses.
Participants who no longer are in the organization are in square brackets.

Upcoming Meetings: The bridge numbers for future IBIS teleconferences are as
follows:

  Date Bridge Number Reservation # Passcode
  February 16, 2001 (916) 356-9200 2-508083 4216977
  March 2, 2001 (916) 356-9200 2-508084 6782941
  March 16, 2001 European IBIS Summit Meeting - No Phone Bridge

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 AND MEETING QUORUM
The IBIS Summit Meeting was held in Santa Clara, California at the Santa Clara
Convention Center. About 72 people representing 35 organizations attended.
The notes below capture some of the content and discussions. The meeting
presentations and other material will be uploaded at:

  http://www.eda.org/pub/ibis/summits/jan01/

Bob Ross opened the meeting by asking everyone to introduce themselves. The
group was well represented by Semiconductor and EDA tool vendors and users of
IBIS models. Also several people from EDA commercial IBIS model providers
and EDA consultants attended. Several people were also involved in other
official standards bodies, some of which are considering using IBIS models
for simulation models.

Bob thanked DesignCon for providing the meeting room, refreshments and the
Booth as part of our Associate Sponsorship of DesignCon. The IBIS Open Forum
is mentioned in the literature. Bob thanked National Semiconductor for
providing the hot lunch and Milt Schwartz for taking care of the local
arrangements including providing copies of the presentations. Bob thanked
Applied Simulation Technology for loaning an LCD projector. Bob thanked
Guy de Burgh and Innoveda for taking care of the booth backdrop and handling
the member signs. The IBIS Booth is #861 on the show floor. Finally, Bob
thanked the presenters and participants for attending.

IBIS TODAY
Bob Ross, Mentor Graphics
Bob Ross briefly noted the current IBIS activity:

  Ongoing Version 3.2 activity including ibischk3 Version 3.2.6 and more
    BUG processing.
  IBIS Version 4.0 BIRDs being processed
  Connector Specification (Version 0.960 has been uploaded)
  Future "IBIS-X"

We are also tracking other activities including EIAJ IMIC progress, JEDEC
activity and European TC93 WG6 EMC/EMI activity. We are also noting that
there may impacts from other standards.

The meeting today progresses from present to futures:

  Current IBIS Model Development and Validation
  Future Needs
  Connector Specification Progress and Discussion
  Future IBIS-X Progress and Discussion
  Other Topics and Discussions

Bob then asked for any new topics. Ian Dodd's report on HTML format for IBIS
will included as part of another presentation.

IBIS DOCUMENTATION FOR ACME ENGINEERING
Scott McMorrow, Siqual Corporation
Because of equipment availability, Scott McMorrow started first. He stated
that his presentation was a reduced form of an actual report on a model he
developed for a customer (with names changed). It shows a process for
developing and validating IBIS models.

The first problem is gathering all of the specification and Spice model
information. Sometimes initial information is incomplete, so this process
may take several weeks. The IBIS model are developed by using internal tools
based on improvements to the North Carolina State University s2ibis2 utility.
Once completed, the model is checked against several test circuits similar
to those in the IBIS Accuracy Report. In particular a transmission line with
a small capacitive loads are used. The Figures of merit similar to those
in the Report are given with the model. Scott expects values in the range
of 97% to 99% for released models.

Scott mentioned that the major EDA simulators that process IBIS models give
essentially the same results. He showed examples of IBIS buffer simulation
overlayed with Spice model simulation. Some differences were based on DC
errors in the I-V data. Sometimes waveform mismatch between some expected
DC starting and ending values and those documented in the waveform tables
caused some overshoot glitches. The other main source of mismatch were
related to the number of points in the V-T tables. IBIS has an overall 100
point limit that sometimes must be allocated between the different time
shifted typ, and max columns. He had observed that with 33 points he may
get 6% error in correlation. Scott indicated that, if necessary, more
points can be extracted and 100 points per column and can achieve 99%
correlation. With 256 points, correlation is excellent and delays are
with picoseconds.

Scott extracts the die capacitance from the model. A wrong die capacitance
provides the second largest source of error, especially at the receiver
where the reflected amplitude is doubled. He described two methods. One
is based on the response to a current source. The other is based on
simulation of a driver and receiver and use Spice to find the value through
optimization.

Scott illustrated these points by showing overlaying Spice simulation and
IBIS model simulation examples and noting some of the mismatch details and
the probable reasons for the deviations.

Scott agreed with a comment from a participant that the correlation metric
number itself could depend on the length of the period (and the corresponding
amount of overlaying data). Visual inspection for differences should also
be done. Scott concluded that IBIS model simulation can be as accurate as
Spice model simulation. Other variables such as actual board impedances
contribute far more deviation in actual measurement to simulated results than
the accuracy deviations discussed in the correlation study.

                                                                      
THE JOY OF WEB MODELS
Tom Dagostino, Mentor Graphics
A customer had asked Tom Dagostino to "QC" (do a quality control check) a set
of IBIS models needed for a project. About 20 of them were obtained directly
from semiconductor web sites. Most actually were quite acceptable and may
have needed only minor, easy to do addition or corrections such as adding
threshold and timing values to some models. However, Tom had three really
bad examples to share. The model names and vendor names were changed to
protect the guilty!

While the review process involves more steps, the models illustrated in the
presentation failed either or both the ibischk3 and visual inspection check.
The first model was a clock driver that had serious ibischk3 mismatches for
rising and falling waveforms and I-V data. Visual inspection revealed that
the pulldown table had zero current in the active region, as the source to
the problem. A visual inspection also revealed that C_comp value was
constant for all three cases: typ, min, and max. Tom reported these problems
to the manufacturer, and after the second call he received a corrected model
that was satisfactory.

The second model was documented as a SCSI terminator. It passed ibischk3
with zero errors. However the model type was listed as Output complete
with [Pulldown] and [Pullup] tables and [Ramp] data. The manufacturer has
been contacted and stated that they would pass the information on. However,
the manufacturer has not responded.

The third component had at least several bad models. A model named as an
_in model with a signal_name and Model_type Output had ibischk3 waveform
errors. Closer investigation showed a number of interesting aspects: Vinh
and Vinl were defined; the typ, min, max temperatures were not correct; and
the I-V table data contained zeros in the normal operating regions. Also,
the [Ramp] data contained zeros. Another _in model was documented as an I/O
model with weak [Pullup] and [Pulldown] data and more [Ramp] data problems
including negative time. Another had a constant voltage [Falling Waveform]
table. A model, really documented as an _out model, had meaningless and
unrealistic conditions regarding the content of the data: a "rough
approximation" to lab measurements, and a max condition that was not
consistent with the IBIS definitions. It passed ibischk3, but with waveform
warnings. One source of warning was a repeated voltage value. Some tables
had strange values inconsistent with typ, min, and max values. This time the
[Pulldown] table had -0A values and [Ramp] data values of 0. For this
component, the manufacturer has not yet responded to Tom's feedback.

Tom concluded that some model developers need to do a better job in making
models and QC'ing them before they are released.

LVDS MODELING
Hazem Hegazy and Mohammed Korany, Mentor Graphics
Mohammed Korany reported on the problems associated with creating IBIS models
for LVDS differential driver buffers with internal differential terminations.
The basic problem is to extract the proper DC I-V data. He reviewed several
existing methods and then two new proposals. The last proposal gave the best
results.

First Mohammed reviewed LVDS. He then reviewed two methods that have been
proposed. The first one dealt with a voltage sweep at PAD with PADN
connected to R_load and Vref. The second method used a mirrored negative
current source at PADN. A third method was also presented where mirrored
delta voltages are applied at PAD and PADN. All of these methods worked for
50 ohm loads and provided nearly overlaying IBIS and Spice simulations. Some
DC offset was observed with 100 ohm terminations. However, all methods gave
DC offsets when internal differential terminations existed in the LVDS model.
Mohammed accounted for these offsets by showing the mathematical distribution
of currents during extraction of the model.

Proposal I was similar to Method 3, but had the voltages attached to a Vmid
value derived from averaging the low and high state voltages. This produced
better results but still showed DC differences with internal terminators and
the load equal to 100 ohms. It only supported the using the R_fix value
load.

Proposal II used mirrored resistance changes with resistors connected to the
Vref value. This produced overlaying simulations for all cases. Mohammed
showed overlaying simulations with a test circuit where the differential,
internally terminated buffer was driving a lossy, coupled differential line
into a receiver load. The results overlaied well for all differential
conditions.

Mohammed noted that one could remove the internal differential resistor as
another approach. Bob Ross noted that we may not know exactly the internal
resistor (if the driver does not have a high Z or input mode or if the Spice
model is encrypted).

Several people noted that the examples showed larger differential voltage
swings than specified for LVDS. Mohammed stated that this was from an ASIC
produced by one vendor. The methods discussed are applicable to LVDS.

Arpad Muranyi (several methods were referenced from his class material)
stated that this was good information. He never did consider the impact of
internal differential terminations.

Ed Sayre stated that even other terminations including AC terminations may
exist. It is important to get the model correct for DC and AC responses.

Mohammed concluded that LVDS IBIS modeling needs high DC resolution

LSI POWER AND GROUND MODEL FOR EMI SIMULATION
Norio Matsui, Applied Simulation Technology and Hiroshi Wabuka, NEC
Norio Matsui introduced the problem by showing several simulation comparisons
between IBIS Simulation, table Spice (IMIC) simulation and Spice (all done
using the Applied Simulation Technology tool. Standard simulation of LVDS
and PECL circuits work well. However, IBIS loses some detail with internal
inverter circuits and also does not produce the same simultaneous switching
output (SSO) responses for currents and voltages. ICs with complex internal
structures cause unaccounted internal currents that need to be added to the
model for ground and power modeling and EMI.

For large scale integration (LSI), the conventional IBIS structures can be
used for input and output. An internal current circuit is added between the
input and output structures. Such an internal structure might represent a
clock macro. The techniques for macro modeling can be:

  I/O IBIS or IMIC for I/O + Package
  Clock 10% consumes most power (highest frequency)
  Non Clock Behaves a filters/decoupling capacitors

Norio described the procedures for macro modeling and showed some Spice
and table Spice structures and also some IBIS EBD and IBIS I/O structures.
The clock and non-clock model can be realized using a cascaded double IBIS
I/O for the dynamic portion and a set of RC networks for the non-clock static
portion. Norio showed a final circuit containing double inverters for the
middle stage.

Using this approach, Norio showed accurate correlation between the IBIS/EBD
and table Spice (IMIC) models for current waveforms. There was some loss in
accuracy when he simplified the model using only four IMIC transistors. Norio
also showed accurate comparisons simulated and measured peaks of near field
EMI.

Arpad Muranyi questions whether splitting C_Comp into components to different
rails might improve the accuracy for the simplified model. Also die
interconnect details were not simulated in this study.

SCSI COMPENSATION MODELING
Larry Barnes, LSI Logic
Larry Barnes introduced himself as co-chair and technical editor of a Ad hoc
Working Group involved in T10. T10 is a Technical Committee of the National
Committee for Information Technology Standards (NCITS). It was formerly
known as the X3 Committee. NCITS is accredited by and operates under rules
approved by ANSI. Larry showed the membership and gave some background on
the Small computer Systems Interface (SCSI).

A technical synopsis of SCSI configurations include:

  Bus structure with 27 data, 8 ground, 4 termination power, 2 reserved lines
  68 conductors, 90 - 120 ohms, single ended or differential
  25 meters point-to-point
  12.5 meters with 16 stubs
  Allowable loading of 30 pF / stub

Larry presented a number of issues that produce data errors associated with
320 MBytes/second transfer rates, decreased setup and hold time, intersymbol
interference, decreased cable charge time, and 16 capacitively loaded stubs.
The eye diagram without compensation has decreased, and the worse case
situation has no opening.

Larry proposes a solution which produces a very acceptable eye diagram:

  Run length encoded compensation at the transmitter
    Transmitter overdrives unless two or more data bits are sequentially
      the same value
    Second data bit and subsequent bits of the same value have the drive
      level reduced
    Drive strength, rise time and fallback are programmable
  Training pattern upon initialization with bit rate negotiation

Larry is writing an ANSI/NCITS Technical Report on SCSI Signal Modeling to
document the details of the solution. The Working Group has selected IBIS a
the data exchange format for:

  Semiconductor device models
  Termination models
  Connector models (when codified)

The Working Group would like to continue to use IBIS if fallback modeling
can be accomplished. Currently IBIS Version 3.2 can be used for SCSI devices
for:

  Single-ended models
  High voltage differential models
  Low voltage differential models

SCSI 320 Mbyte/s driver characteristics require a fallback model:

  Differential current mode transmitters
  Negotiable data rates
  Multi-level parallel positive and negative current drivers
  Uses programmable fallback

Fallback cannot be modeled in IBIS Version 3.2:

  Driver schedule keyword is close
  Needs bit position dependency, not time
  Future may include overdrive past first bit
  Needs accommodation for parallel current mode drivers

Larry proposes adding a [Fallback Schedule] keyword:

  Similar in construction to [Driver Schedule]
  Describes switching sequence fallback modeling
  Delay parameters are in terms of bit times
  Accommodates typ, min, max conditions
  Differential operation
  Two current sources and sinks for each line

This will cover the next generation of SCSI and IBIS futures through 2003.
Subsequent generations may include phase encoding. Larry hopes that this
will fit in the future of IBIS.

Larry welcomes IBIS participation in this Working Group. A draft work in
progress specification is available at:

  http://www.t10.org

LUNCH
The Group broke for a delicious buffet lunch sponsored by National
Semiconductor.

CONNECTOR SPECIFICATION
Gus Panella, Molex
Gus Panella reported on the progress of the IBIS Connector Specification.
There have been some delay because of the loss of a key member (Kellee
Crisafulli), limited participation, holidays, and working with a big
document. However, the Working Group has been improving the descriptions
and clarifying ambiguous syntax.

Gus reported on some of the changes:

  Removal of Physical Map because of unnecessary complexity
  Distinction of Side_A and Side_B pin maps
  Clarification of [Redistribution] Specific to mean comments are in the
    Header section
  Some number of pins, rows, and columns ranges to document physical limits
  More [Cn_Row_Swath] and [Cn_Column_Swath] distinction and edge definitions
    and usage

Gus reported that Ian Dodd provided a C-style syntax to describe the pin
map information as a function of ROW and COLUMN entry specified by the EDA
tool. Ian gave the brief description (given below) at this time.

Gus stated that the current discussions involve following the IBIS-X group
header suggestions, where practical. These include bracketing the header
with [Begin Header] and [End Header] keywords. Gus asked if the [Source]
keyword should be required. It is not required in all other documents. The
consensus was mostly in favor on not making it required after some meeting
discussion. Gus also asked about using " " (space) or "_" (underscore)
between keywords, but this was clarified as an documentation editorial detail,
not a functional detail. Both conventions will be supported. The work in
progress (Version 0.960) of the Connector Specification has been uploaded for
review.

Gus closed by mentioning some short term goals. He concluded that the
document is reaching technical completion. Bob Ross added that he plans to
release the document for full IBIS Committee technical review and comments
while the editorial cleanup of the document is being done. The document will
go through formal review processes for EIA IBIS Open Forum approval, and
further review processes (including a 90 day formal public review) for EIA
ratification.

IBIS CONNECTOR SPECIFICATION: PRELIMINARY PROPOSAL FOR A PIN NAMING LANGUAGE
Ian Dodd, Cadence Design Systems
As part of the Connector Specification presentation, Ian Dodd displayed and
discussed aspects of a proposal for pin naming language. The main points of
the proposal are:

  It adopts C - like code style including C style formatting statements
  It predefines certain variables ROW, COLUMN, SIGNAL_NAME, and PIN_NAME
  It presumes all variables are integers and do not need to be declared
  It has some simple math operations and conditionals
  It can easily specify incrementing characters through addition

Ian showed some applications including one where a row was missing. He
stated that the proposal might be compatible with the IBIS-X macro language.
One issue is whether they should be aligned. The text formatting might be
needed for the macro language. One pending decision is whether the C style
of ending lines with semicolons should be done in the pin naming language.

AN API FOR IBIS??
Al Davis, Consultant
Al Davis gave an informative presentation on an application programmable
interface (API) for IBIS. He defined an API as:

  A binary interface for models
  It supports portable models external to the simulator
  It provides the ultimate generality and intellectual property protection

Al listed a number of issues:

  Type of interface
  Operating System differences
  Language differences
  Simulator differences
  Uses beyond simulation
  Hard to make portable (easy for one simulator)

The interface can be dynamically linked object modules (faster, but harder
to develop and less portable) or a separate application or applet (slower,
but more general and portable). Al prefers the executable approach. He
listed some advantages and issues. It should support different types of
simulators ranging from Spice-like, mixed-mode, transmission line / tree
based, gate level, and HDL. It should also support more than one simulation.
There was some discussion on whether API's would be repeated if the model
was used several times. This needs to be clarified.

The functions needed for an API are:

  Parsing and printing the calling netlist
  Pre-processing
  Predictor step
  Evaluation
  Gather and scatter
  Review step (iteration and step control)
  Probes
  Status queries
  Plus about a dozen more

Normally the user would use other tools to write the functions including
VHDL-AMS, IBIS-X, etc. The simulator developers would do the work since
the complete understanding of simulator internals is needed.

Al concludes that an API is needed, but it will take coordination between
EDA companies. It has potential beyond IBIS.

IBIS FUTURES GROUP UPDATE ON IBIS-X
Stephen Peters, Intel Corporation
Stephen Peters, who is serving as Chair or the IBIS Futures Working Group
introduced the progress on IBIS-X. He listed the goals:

  Provide extensible syntax
  Enable complex/coupled package descriptions
  Maintain backwards compatibility (IBIS Version 3.2)
  Support some additional requirements (equation based modeling, HTML viewing,
    public key encryption)

Briefly, IBIS-X provides the following:

  Nodal based package description (die interconnect pin to pad, power
    delivery, and pin coupling)
  Macro language for creating new model types

The current status is:

  Al Davis has implemented IBIS Version 3.2 in a macro language
  The Working Group meets bi-weekly on writing a formal specification:
    Overall IBIS-X Specification
    Library Guide
    Programmers Language Reference Manual (LRM) expected in months
  Work remains on the die interconnection section

Stephen introduced some current issues:

  Alignment with the Connector Specification
  Case sensitivity in .ibs files (need to create rules for capitalization
    if it is supported.
  HTML support
  Support of more general equations and black boxes such as the Berkeley
    Spice B element
  Where does the API fit into this?

(Later Al Davis raised the case sensitivity issue with the group. The
majority indicated preference for removing the requirement.)

PROPOSED NEW DIRECTIONS FOR IBIS PCB SIGNAL INTEGRITY MODELS
Ian Dodd, Cadence Design Systems
Ian Dodd provided an overview of IBIS today and future needs. He gave the
brief history of IBIS development and features and described its well-known
modeling advantages. He also listed some limitations:

  Feature changes require the slow process of adding new keywords
  The rate of requests for feature changes is accelerating
  Models assume a fixed topology
  The present standard is inadequate for todays needs in:

    Input transition detection
    Simultaneous switching noise
    Power delivery
    Coupling within packages

Ian listed a number of goals, similar to those discussed by Stephen Peters.
IBIS-X for the next generation should:

  Extend the data sheet
  Add a new structure template

Ian illustrated the extended data sheet (similar to IBIS) and noted the
structure details and topology. He also listed a number of circuit elements
for IBIS-X

Ian also discussed and illustrated how IBIS and IBIS-X can be made compatible
with HTML. A feature is that lines with "<" in column one are assumed to be
HTML formatting commands. Some details have to be worked out regarding some
initial declarations. HTML can also be used to add separate graphics file
pictures within IBIS files. Ian illustrated doing this using a portion of
an example provide by Stephen Nolan. Scott McMorrow raised the concern that
the Working Group might be working on fluff and not dealing with the critical
issues. Ian responded that this issue was really not taking much time.

Ian listed and discussed a number of advantages of IBIS-X:

  Extensible
  Allows reuse of structural template
  Data is separated from templates
  Devices in a wide range of logic families can share structural templates
  Can read existing IBIS models
  Nodal structure allows multiple levels of abstraction

He also listed some concerns:

  High expertise is needed to create new IBIS-X templates
  Model developers will need to create more than one type of data sheet
  Simulation tools will have to be more intelligent in syntax checking.
  Model validation will be more difficult.
  
Ian listed the future IBIS-X work and noted that there are still some open
issues and concerns regarding the level of abstraction and whether exiting
standards can be used. Ian concluded that IBIS-X is promising, but it will
require increased sophistication of validation tools and training for model
developers.

IBIS-X PROGRESS REPORT
Al Davis, Consultant
Al Davis listed the progress:

  Some documents
  Working parser
    Tested with about 100 IBIS files
    Some bugs
    Generates partial Spice deck

Al's goals are

  Doing simulations by March
  Also include the Connector Specification
  Produce a presentable version of the Specification by March

He currently has a working demonstration. It can be downloaded (pre-alpha
code for review). It can be found at:

  http://table.jps.net/~atd/ibisx.200101.tar.gz

IBIS-X MACROLANGUAGE
Al Davis, Consultant
Al Davis reviewed a previous presentation on IBIS X. He stated that the
basic simulator requirements were:

  Handling piece-wise linear (PWL) constructs
  Handling two-dimensional PWL tables (time and signal)
  Having a "trigger" to shift time tables

Additional elements include generalized expressions for

  Voltage, current, charge, flux
  Multi-port block

IBIS-X supports structure (a spice like description with expressions). IBIS
already contains data with data sheet like information for tables and
attributes. These were illustrated.

Other structures Al illustrated were conditionals (if), time dependent tables
and triggers, an IBIS compatible driver element, a new foreach statement, and
an alarm, and value expression.

A number of points were discussed including having a floating gnd versus
absolute 0 voltage ground and having $ before variables.

CONCLUDING ITEMS
Bob Ross again thanked the sponsors of the meeting and also the presenters
and participants.

Bob stated that the next teleconference meeting is planned for Friday,
February 16, 2001, and the following one on March 2, 2001. Actions are still
needed for issuing BIRD65.2 and BIRD68.1 prior to voting on these.

Bob reminded us that the European IBIS Summit Meeting is scheduled for
Friday, March 16, 2001 in Munich Germany in a hotel near the Design Automation
And Testing in Europe DATE 2001) conference and exhibition.
The four co-sponsors are Cadence Design, Innoveda, Mentor Graphics, and Zuken
(Incases). The first notice was sent out at the end of January 2001. So
far Bob has about three or four presentations proposed so far, so he expects
this to be an all day meeting with a free lunch to participants.

Bob asked if there were any questions or issues to discuss briefly. Kim
Helliwell asked about what was happening to the s2ibis3. Michael Cohen
commented that after much debate we decided not to seek funding on the
project because it would create a product that already competed with a
commercial product. Bob added that there were other commercial modeling
services as well. However, the project specification is still uploaded, and
anyone can still work on it.

NEXT MEETING:
The next teleconference meeting will be on Friday, February 16, 2001 from
8:00 AM to 10:00 AM. BIRD65.2 and BIRD68.1 are scheduled for a vote if
ready.
==============================================================================
                                      NOTES

IBIS CHAIR: Bob Ross (503) 685-0732, Fax (503) 685-4897
            bob_ross@mentor.com
            Modeling Engineer, Mentor Graphics
            8005 S.W. Boeckman Road, Wilsonville, OR 97070

VICE CHAIR: Stephen Peters (503) 264-4108, Fax: (503) 264-4515
            stephen.peters@intel.com
            Senior Hardware Engineer, Intel Corporation
            M/S JF1-209
            2111 NE 25th Ave.
            Hillsboro, OR 97124-5961

SECRETARY: Guy de Burgh (805) 988-8250, Fax: (805) 988-8259
            gdeburgh@innoveda.com
            Senior Manager, Innoveda
            1369 Del Norte Rd.
            Camarillo, CA 93010-8437

LIBRARIAN: Roy Leventhal (837) 797-2152, Fax: (847) 222-2799
            roy_leventhal@3com.com
            Senior Engineer, CommWorks Corp. (a wholly owned 3Com subsidiary)
            1800 W. Central Rd.
            Mt. Prospect, IL 60056-2293

WEBMASTER: Syed Huq (408) 525-3399, Fax: (408) 526-5504
            shuq@cisco.com
            Manager, Hardware Engineering, Cisco Systems
            170 West Tasman Drive
            San Jose, CA 95134-1706

POSTMASTER: John Angulo (425) 869-2320, Fax: (425) 881-1008
            jangulo@innoveda.com
            Development Engineer, Innoveda
            14715 N.E. 95th Street, Suite 200
            Redmond, WA 98052

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@eda.org
      To join, change, or drop from either the IBIS Open Forum Reflector
      (ibis@eda.org), the IBIS Users' Group Reflector (ibis-users@eda.org)
      or both. State your request.

  ibis-info@eda.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@eda.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. Job posting information is not permitted.

  ibis-users@eda.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. Job posting information is not permitted.

  ibischk-bug@eda.org
      To report ibischk2/3 parser bugs. The Bug Report Form Resides on
      eda.org in /pub/ibis/bugs/ibischk/bugform.txt along with reported bugs.

      To report s2ibis, s2ibis2 and s2iplt bugs, use the Bug Report Forms
      which reside under eda.org in /pub/ibis/bugs/s2ibis/bugs2i.txt,
      /pub/ibis/bugs/s2ibis2/bugs2i2.txt, & /pub/ibis/bugs/s2iplt/bugsplt.txt
      respectively.

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.eigroup.org/ibis/ibis.htm

Check the pub/ibis directory on eda.org for more information on previous
discussions and results. You can get on via FTP anonymous.
==============================================================================

 
Received on Fri Feb 2 14:49:51 2001