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

IBIS INTERCONNECT TASK GROUP
https://urldefense.proofpoint.com/v2/url?u=http-3A__www.ibis.org_interconnect-5Fwip_&d=DwIGAg&c=nKjWec2b6R0mOyPaz7xtfQ&r=DcQR-qLpQg5lIreuM6-NYECRIAFXt268PRNS5WO043M&m=aUWePdXi5y9929TG6cP2hWDnE_kSdebBuOqU8nhDE7zPii64q1TN9rpzYz7kSy5G&s=4amX5JpUvB4NzNn3lnLkuxLjgbQxFKV4EOatV5SA5kw&e=  
Mailing list: ibis-interconnect@freelists.org 
Archives at https://urldefense.proofpoint.com/v2/url?u=http-3A__www.freelists.org_archive_ibis-2Dinterconn_&d=DwIGAg&c=nKjWec2b6R0mOyPaz7xtfQ&r=DcQR-qLpQg5lIreuM6-NYECRIAFXt268PRNS5WO043M&m=aUWePdXi5y9929TG6cP2hWDnE_kSdebBuOqU8nhDE7zPii64q1TN9rpzYz7kSy5G&s=GxPGOFkf1L9Ovmlp7HFt1PUsJcGbNwzqyJLG_MwyU8k&e=  

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

Attendees from April 20, 2022 Meeting (* means attended at least using audio)

ANSYS                                Curtis Clark
Intel Corp.                          Michael Mirmak*
                                     Michael Brownell
Marvell                              Steve Parker
MathWorks                            Walter Katz*
Micron Technology                    Justin Butterfield*
                                     Randy Wolff*
Siemens EDA                          Arpad Muranyi*
ST Microelectronics                  Aurora Sanna
Teraspeed Labs                       Bob Ross*
University of Illinois               Jose Schutt-Aine
Zuken USA                            Lance Wang*


Michael Mirmak convened the meeting.  No patents were declared. 
Justin Butterfield and Randy Wolff took minutes.


Review of Minutes:
- Michael called for review of the minutes from the April 13, 2022 meeting.  
  Michael displayed the minutes.  Arpad Muranyi noted there is a typo, where 
  "look out" should be "look at".  Michael noted "foster" should be 
  capitalized, and he suggested "element" should be "controlled source".  Arpad 
  motioned to approve the minutes with these changes.  Randy seconded.  The 
  minutes were approved without objection.


Review of ARs:
- Arpad to convert Vladimir Dmitriev-Zdorov's format to a keyword based format 
  and start a BIRD draft.
  - Arpad noted the draft is started.  He has not sent it out, as he still has 
    some questions for Vladimir.


Opens:
- None.


Discussion:

Port Naming:

Walter Katz added discussion topics to the port naming slides.  On the logical 
and physical names, the request from Cadence was to have both.  On the 
reference information, the request was to include the reference information.  
He did not get a response on Chord signaling.  Michael asked if there are any 
papers discussion the Chord signaling.  Arpad noted C-PHY is the common example 
with 3 wires.

Walter noted the reference information could be an additional section with the 
physical location of the reference.  It could have a Net name and Physical name 
of the reference.  

Michael asked if it is worthwhile to split the end and net into their own 
keyword.  In some cases the end information would be the only information of 
interest.  Walter suggested the end and net are critical information and the 
order of the columns could be changed.  

Arpad asked if we should rename "end" to "side".  Walter noted we could call it 
"side", but you may have more than two sides.

Bob Ross asked why we need multiple locations for the referencing.  Arpad 
replied that we could have grouped or merged pins for the reference.

Walter proposed an additional format to replace differential with Chord.  Arpad 
suggested to not have different columns for the Chord information.  We need to 
find a way to identify groups with a combined Chord/differential notation.  

Michael asked Walter if he would send out the presentation with changes made 
during the meeting.  Walter to send out the updated presentation [AR].

Bob asked if we need to support Chord.  He was concerned we were making it too 
complex and may still be leaving out possible technologies to support.


PLS in Touchstone TSIRD Draft:

Arpad shared a draft TSIRD for inclusion of pole/residue data in Touchstone.  
In the proposed changes section, he references Vladimir's presentation and 
graphics from his slides.

The existing option line in Touchstone could be reused.

For pole/residue data, three keyword pairs are needed.  [Begin Pole/Residue 
Data] is a complete set with common poles and residues.  The [Begin Common 
Poles] and [Begin Residues Data] sections could be used for common poles with 
residues at different frequencies.  Delay and Asymptote are optional 
parameters.  

The original [Mixed Mode Order] keyword could be used with the new data but 
details are not added yet.

Arpad noted the Number_of_Poles parameter may not be needed.  
Number_of_Residues needs to match the Number_of_Poles.  For residues, the 
Delay parameter is not allowed.  

Arpad is waiting for Vladimir to confirm if the Asymptote parameter is 
permitted for H and G-parameters.

Walter asked, for a 4x4 with 16 sections, what happens if one is missing?  Is 
that interpreted as zero for that i,j value?  We should think about how to do 
a sparse matrix.

Walter commented a way of indicating that two sets of data are the same, the 
syntax could use the following for swathing:
[Begin Pole/Residue Data] (8,8) = (1,1)

Randy asked about support for complex poles.  Bob noted the examples shown are 
complex values in two columns.  Complex conjugates are assumed.  

Bob commented he would support a separate specification but would support this 
specification combined with the standard Touchstone specification.  Combining 
together can make a tangled set of rules of what is required.  

Arpad noted the problem is the Touchstone files are huge.  The pole/residue 
data could be at the beginning of the file along with the original Touchstone 
data.  The simulator could read the start of the file and ignore the rest.  
There are pros and cons of separate files or combined files.  There is a lot 
of common information including the port naming syntax, etc.  Bob noted port 
mapping could be an optional capability in either format.  

Bob said, with the format shown, we can easily make reciprocal data, where 
Si,j = Sj,i and not duplicate data.

Arpad asked if the format Walter had shown really applied to swathing of a 
connector.  Walter noted swathing implies the same values everywhere, whereas 
his example was applying the same values where possible.


Arpad motioned to adjourn.  Bob seconded.


Next Meeting:
The next meeting will be April 27.


================================================================================
Bin List:
1. Touchstone 3
2. Pole-residue support for Touchstone
3. Port naming