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

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

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

Attendees from December 3, 2025 Meeting (* means attended at least using audio)

ANSYS                                Curtis Clark, Wei-hsing Huang, Juliano Mologni
Arista Networks			     Jim Antonellis
Broadcom                             James Church
Chipletz			     Stephen Newberry
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.

Michael reviewed three previous sets of meeting minutes that had been distributed
earlier to the reflector.

For the September 24 meeting, Arpad Muranyi moved to approve the minutes; 
Randy Wolff seconded.  The minutes were approved without objection.

For the October 29 meeting, Randy moved to approve the minutes; Arpad seconded.  
The minutes were approved without objection.

For the November 12 meeting, Arpad moved to approve the minutes; Walter Katz 
seconded. The minutes were approved without objection.

Michael reviewed a reflector note he had sent explaining the usage of "node"
and "terminal" in Touchstone and IBIS.  This investigation was the only AR from
the previous meeting.  As a result, he was replacing "node" with "terminal" 
throughout the Port Mapping draft.

Arpad replied that he had no objection to using "terminal", but wants to 
avoid a blanket proscription on the use of "node".  Weston Beal added that it
would be useful to include a formal definition; some terminals may become nodes 
in future usage or contexts.

Arpad stated that his previous internet searches showed "terminal" was defined
as a place to make connections, with "node" being a name for the point after 
the connection is made.  Terminal is an interface point for a black box circuit. 
Weston agreed with the "one point of interface" to a circuit being a terminal.

Michael added that the team was not going to address any required IBIS changes 
or investigate IBIS-ISS usage (the latter specification was not reviewed for this
AR).  

During review of the Port Mapping draft, Michael raised several questions related
to the use of "+" and "-" symbols.

1) The "+" symbol is to be required?
2) Should "Physical" be required?  Alternately, is an all-Logical case acceptable?
3) Confirm that "Reference" and variants should be removed entirely in favor of "-"?
4) Does the current syntax support the GSSG probe structure?

Regarding whether the "+" symbol should be required, Arpad suggested that the "+" 
sign should be optional.  Michael asked whether it is implied everywhere.  Where 
would it not make sense?

Arpad asked whether more than one set of parentheses would be supported.  If so,
can they be nested?

Michael referred to a new syntax format example in the draft:

   Port n +(name value name value...) -(name value name value...)  (name value)

Walter suggested that "-" only be used in front of reference terminal information.
This would resemble:

   -(Logical blah) -(Physical blah) 

 and only for those two name-value pairs.  He prefers requiring the "+" symbol.

He added that most pin mapping cases will not include any reference 
data at all.  This means that ports and terminals could default to positive. 

Michael stated that, in the current draft, the "+" and "-" characters are 
only required when duplication of terminals is present.  Arpad asked whether
the team was now removing the Reference_* syntax.  Michael confirmed this.

Michael recalled earlier proposals from Arpad with the "+" and "-" symbols.
Arpad suggested that there only "-" was required.  Walter agreed that using 
"-" is acceptable.

Arpad asked whether repeated parentheses should be allowed.

Walter stated that he believed "+" should apply to the port.  Weston and 
Arpad disagreed.

Walter added that what is differential is the positive terminal; the negative is not 
differential.  Michael reminded the team that Diff_port is not associated 
with polarities.

Walter stated that he does not measure by probing on the negative terminals.

Weston replied this is the result of long-standing omission in industry of 
reference terminals in measurements, simulations, and other contexts.

Arpad: two ports are defined in terms of mathematics 1p - 2p
The port is not the + terminal; it's the combination of the terminals.

Walter proposed a two-port structure with four terminals: a+, a+return, a-, a-return

Randy noted that he was researching GSSG probes, looking at pictures to see if two grounds are 
connected together.  Not immediately clear.  One has 2 coaxial connections, 
indicates that "ground" only goes to one coaxial cable.  Probably end up 
with different reference for the two signals.  

Walter stated that the extra-parentheses syntax is harder to read.  Arpad added
that the examples are missing multiple "+" symbols.

The team examined one of the early examples:

Port 1 (Physical U7.1)
Port 2 (Type S) (Physical U7.2)

... and experimented with three variations on potential syntax.

Port 3 (Physical U7.4) (Logical DQ5+) -(Physical U7.3)  (Diff_port 4) 
Port 4 (Physical U7.5) (Logical DQ5-)  -(Physical U7.6) (Diff_port 3)

Port 3 +(Physical U7.4) +(Logical DQ5+) -(Physical U7.3)  (Diff_port 4) 
Port 4 +(Physical U7.5) +(Logical DQ5-)  -(Physical U7.6) (Diff_port 3)

Port 3 +((Physical U7.4) (Logical DQ5+))  (Diff_port 4)  -(Physical U7.3)
Port 4 +((Physical U7.5) (Logical DQ5-))   (Diff_port 3)  -(Physical U7.6) 

Randy added that a GSSG probe would use the terminal at U7.6.

Arpad stated that he personally prefers (( grouping, as multiple "+" and "-" 
symbols imply multiple references.

Michael confirmed that the "-" symbol is required if your reference is not A_gnd.
Michael also confirmed with the team that the use of "-" implies that the 
"Reference" term is removed from the draft.  Randy replied that he prefers
the "-" symbol to "Reference".

Arpad asked whether it should be legal to assign "+" to A_gnd.

Randy replied that one cannot measure A_gnd on a VNA.  Michael asked whether 
we want to prohibit this syntax in the proposal.  Randy suggested that one
would not describe the measured data as using "+A_gnd"; one can flip the 
terminals in actual usage.

Arpad stated that this will need rules; Randy replied that these were
simple rules to write.  Weston asked whether the draft previously contained 
a statement that Logical or Physical is required.

Walter replied that one should be able to use Logical, Physical, both or 
neither for a given Port.  One could have an S2P file with Sides, Nets, etc.

Michael took the AR to assemble one such example as part of his overall 
cleanup of the existing examples. [AR] 

Walter noted that Touchstone files are part of the new PIM BIRD being 
discussed in IBIS-ATM.  He suggested the team should review it.

Arpad move to adjourn. Weston and Randy seconded.  The meeting
adjourned without objection.

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