================================================================================ IBIS INTERCONNECT TASK GROUP http://www.ibis.org/interconnect_wip/ Mailing list: ibis-interconnect@freelists.org Archives at http://www.freelists.org/archive/ibis-interconn/ ================================================================================ Attendees from September 11, 2019 Meeting (* means attended at least using audio) ANSYS Curtis Clark* Cadence Design Systems Bradley Brim Intel Corp. Michael Mirmak* Keysight Technologies Radek Biernacki Mentor, A Siemens Business Arpad Muranyi* Micron Technology Justin Butterfield*, Randy Wolff* SiSoft Walter Katz*, Mike LaBonte Teraspeed Labs Bob Ross* Michael Mirmak convened the meeting. No patents were declared. Justin Butterfield took minutes. Review of Minutes: - Michael called for review of the minutes from the September 4, 2019 meeting. Randy Wolff moved to approve the minutes. Bob Ross seconded. The minutes were approved without objection. Review of ARs: - Walter Katz to make the correction from "EMD Description" to "Electrical Module Description" in the latest EMD draft. - Walter noted this was done, but Bob found a few more corrections. Walter sent out a draft 18 with these additional corrections. - Walter noted there was an AR in the IBIS Open Forum to discuss Touchstone in the Interconnect meeting. He has prepared a presentation detailing SiSoft's position on enhancements to Touchstone. Opens: - None. EMD Draft 18 Review: Michael noted we have the bin list of comments to go through and Walter has proposed moving to a comment resolution phase. Michael asked if we need to have a formal list of comments. Bob suggested the document is in a preliminary phase, and when we submit the BIRD to the IBIS Open Forum, we can more formally track the comments. Michael agreed. Walter added the [End EMD Pin List] and [End Designator Pin List] keywords. Randy commented that these will need to be added to the keyword hierarchy. Walter replied this is a known issue to be fixed in the next draft. Bob noted the [Pin] keyword does note have an end, but he is okay with this change. Bob asked if we need a [Number of EMD Pins] keyword. Walter replied we could remove this, but suggested for the group to think about it first. Bob commented it is a good cross check to have. Walter stated he believes items 2 and 3 in the bin list are complete. Michael asked if we can review this. Walter replied these are in the latest draft. Walter noted the only issues we need to discuss are numbers 1 and 5 in the bin list. Bob noted another issue is "CAD nets" and "Nyquist" frequency need to be defined. Walter suggested to formally submit these issues. Michael agreed these can be sent out to the reflector. Bob stated he would like to get a good baseline of the document, so we are not logging too many issues in a formal setting. Walter to send out a draft 19 of EMD [AR]. Touchstone Direction: Michael noted there was a suggestion in the IBIS Open Forum to advertise to the SI List the Interconnect Task Group and our work on Touchstone. Michael asked what we should do in the way of a presentation to a broader audience on this subject. Walter shared a presentation on how the SiSoft has approached implementing port mapping for Touchstone models. Touchstone 2 has not caught on, since it is only needed for power integrity. He noted the needs for enhancements include: impedance per port, port information, and sparse matrix. Walter noted SiSoft's approach is to import S-parameter models and process the model. The model is wrapped in a convenient schematic symbol format. A "ports" file is created to define the port names and connectivity. Walter noted the limitations of this approach include not handling: rail ports and multi-drop signal ports. He suggested this could a possible solution, but would need enhancements for the rail ports and multi-drop ports. Walter asked where this information should go. He noted swathing connector Touchstone files is an additional use case to think about. Rational fit is an alternative to Touchstone which defines the transfer function in the IBIS-ISS language and the .ports file information is the same as the Touchstone. For next steps, Walter recommended to start a new Touchstone working group and invite the relevant people. The solution we come up with would need to have a commitment to backward compatibility, by making some of the keywords optional. Bob commented it could be useful to make these keywords optional. Michael suggested we need to raise awareness, then identify the necessary changes. The challenge is to structure this feedback. Walter stated we need an information gathering tour to see what is needed. Michael suggested to create a presentation on the basics of Touchstone and what we have done with the standard. Walter suggested to keep the meetings very structured in terms of presentations and the mailing lists. Michael would like to put together a dedicated presentation in a first meeting, then have a following meeting to invite presentations. Walter stated he would like to hear more from the other EDA vendors. Arpad Muranyi asked how this fits in with the EMD discussions and if we should prioritize EMD over Touchstone. Walter agreed we should finish EMD. Bob was concerned we cannot work on both Touchstone and EMD. Arpad agreed. Randy agreed with the sentiment that we should finish up EMD first. Michael will draft an email response to the SI List to let them know that once we finish current work we can take up Touchstone enhancements [AR]. BIRD198 Discussion: Michael asked if there is any interaction or changes to the IBIS 7.0 Interconnect syntax in the BIRD198. Randy noted the authors have some text around the interaction, but it does not change the Interconnect syntax. He has drafted a response and is waiting for feedback before sending that response. Arpad stated the authors have followed the ATM discussion, so he suggested to keep the topic in ATM. Walter agreed with keeping this as a priority task in ATM. Walter suggested to make things simple for EDA tools. Randy noted there could be cases where both syntaxes could be used for on-die decoupling and package decoupling. Bob stated it is up to the model maker to avoid collisions. The advantage with BIRD198 would be to avoid using BIRD189 and complex IBIS-ISS or Touchstone models. Michael agreed and noted, when interaction questions come up to BIRD189, we can discuss them here. Bob noted he has a Touchstone presentation ready to go. Next Meeting: The next meeting will be September 18. Arpad moved to adjourn. Walter seconded. The meeting adjourned without objection. ================================================================================ Bin List: EMD Comments to be Resolved: 1. Should the [Define Module] keyword be renamed? 2. Documentation of CAD nets, extended nets and signal names definitions. 3. Add bus_labels as possible Terminal_type_qualifiers. - RESOLVED 4. Add [End EMD Pin List], [End Designator Pin List] to keyword hierarchy. - RESOLVED 5. Remove [Number of EMD Pins] keyword? 6. Add definition of "Nyquist" and CAD nets. IBIS-ISS Parser: - IBIS-ISS parser scope document