==============================================================================

IBIS INTERCONNECT TASK GROUP
Mailing list: ibis-interconnect@freelists.org 

================================================================================

Attendees from April 30, 2025 Meeting (* means attended at least using audio)

ANSYS                                Curtis Clark, Wei-hsing Huang, Juliano Mologni
Arista Networks			     Jim Antonellis
Broadcom                             James Church
Intel Corp.                          Michael Mirmak*, Xiaoning Ye
Keysight Technologies                Ming Yan
Marvell                              Steve Parker
MathWorks                            Walter Katz*
Micron Technology                    Justin Butterfield
Siemens EDA                          Weston Beal*, Arpad Muranyi*, Randy Wolff*
Simberian 			     Yuriy Shlepnev
ST Microelectronics                  Aurora Sanna
Synopsys                             Ted Mido, Edna Moreno
University of Illinois               Jose Schutt-Aine

Michael Mirmak convened the meeting.  No patents were declared.

The minutes of the April 23 meeting were not available for review.

During AR review, Michael noted that he had sent out draft 13 of the
port-mapping proposal.

Michael mentioned having held an extended discussion with Pete Pupalaikis regarding 
referencing.  His independent mathematical analysis suggests that the port-mapping
referencing approach is generally correct.  Randy Wolff suggested this might be a 
good IBIS Summit presentation.  Michael will contact Pete on this. [AR]

Michael asked whether an update on IBIS Interconnect activities should be delivered
at the next IBIS Summit.  He took an AR to provide a presentation draft for 
discussion at the next meeting. [AR]

Michael reviewed draft 13 of the port mapping proposal.  Minor edits had been made,
for consistency, but the source file has been changed to "Data_usage", with Ts4file
added and "PI" substituting for "SPIM".  Michael asked whether we should keep PI.  
Arpad Muranyi suggested, also for consistency, removing the word "streamlined" from 
second line of the PI definition.

Arpad suggested changes to Data_usage to make them consistent with keyword usage in
IBIS, including "IBIS_Interconnect_Model" and "EMD_Model".  Weston Beal proposed, 
alternately, to use "C_comp" by itself.  Randy added that the parameter should use "IBIS_" 
as prefix for all of arguments.

Walter Katz suggested, for "PI", using a pointer to a PI structure or method defined in IBIS.  

Randy asked whether, for each of these, whether the document needed a syntax example.  Michael
replied that some examples already exist, but using older parameters.

Arpad replied that more work on the examples is needed.  Randy agreed, adding that, to easily 
read and understand the examples, more edits are necessary.

Michael asked whether the document should present the examples immediately after the syntax
definitions, and for each parameter.  Randy replied that we do that for UD (user-defined).  

Michael noted that IBIS uses at least one example for each parameter in AMI.  Weston stated
that he would like to see examples immediately after each paragraph.

Michael took the AR to add examples after each new instance. [AR]  Arpad suggested that maybe 
the examples for Data_usage should be placed after the Source parameter.

Randy suggestted keeping "side" usage consistent, as the usage is bad in some examples and 
definition text.

Arpad asked whether other shapes should be supported for the "side" approaches, such as triangles 
or polygons.  Weston replied that all these would be informational. Walter opposed the suggestion.

Randy suggested adding text that states the section applies to rectangular symbols.
Michael asked whether some of the parameter arguments are case-sensitive.  Randy replied that
"bus_label" (all lower-case) is the usage.  Also, there is no "Group" in EMD.  The Group should 
be "signal_name".  Arpad noted that this is confusing, as there are overlapping definitions of
signal_name and Group name.  Arpad added that EMD uses "Pin_name"  

Randy responded that, from EMD, there is more text to fix under the "P" ports for rails 
(pin name, signal name, designator name.signal name), as well as *.signal name.  
Michael took the AR to add text from pp. 391-392 of IBIS 7.2 for summary.  [AR]

Randy asked about designator_name in IBIS, as it shows up 3 times in draft 8 on page 408; it 
should just be "designator".  "Group" was removed from IBIS_Interconnect list.

Arpad asked whether we need to explain "bus_label" and signal_name usage issue in Touchstone,
due to the overloaded usage mentioned earlier.

Michael asked whether Ext_ref is universal. Does this need a qualifier?
The team generally agreed that no qualifier was needed.

Michael noted that, while Ts4file edits have been added, we need to check IBIS about how 
referencing is permitted (an ideal node 0 connection is shown there).  Wordsmithing of the 
new Ts4file text in Touchstone is needed.

For PI (SPIM) in Touchstone, the allowed data names were discussed.  Walter suggested that
the text define a group and a list of pins; this is really a cluster.  The text needs to 
provide either a pin name or a group name.  One can define ports on terminals; this would
replace connections on die pads.

Next time, referencing will be discussed, followed by a review of the examples.
Arpad moved to adjourn; Randy seconded.  The meeting adjourned without objection.

The next meeting will be held May 7, 2025.