================================================================================ 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 February 14, 2018 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 February 9 meeting. Mike LaBonte moved to approve the minutes. Arpad Muranyi seconded. The minutes were approved without objection. Review of ARs: - Arpad to draw up a diagram illustrating the potential issue with the Micron decoupling model. - Michael reported this is done. - Bob Ross to describe additional Aggressor_Only rules. - Bob reported that, after further review, additional rules are not necessary. But, a change to the rule on page 27 may be necessary. - Walter Katz to sketch out an example of the on-die decoupling capacitor connecting between the buffer power and ground rail terminals. - Randy Wolff noted this was discussed in ATM, and the underlying question has been resolved. Bob noted there are several solutions to the problem of connecting the on-die decoupling capacitor. Opens: - Mike noted there are several comments in sections of BIRD189.5 that are now deleted. He noted these comments are now marked as resolved. Review from ATM: Mike noted, on page 8, there was a sentence that was changed on Friday, while there was a motion in the ATM meeting to delete that sentence. Arpad stated he prefers the updated version of the sentence. Mike retained the sentence from Friday in the current version of BIRD189.5. BIRD189.5_draft17_v3 review: Mike noted there are some comments on File_TS on page 26. Michael stated that these comments are from Radek regarding his concern that we need to be consistent with BIRD158.7. Michael stated he will clarify the File_TS comments with Radek [AR]. Arpad noted that BIRD158.7 has three drawings each with a ground symbol representing node 0. In BIRD189.5, we do not require that the references connect to A_gnd. Michael asked if this is more of a problem for BIRD158.7. Walter noted that interconnect models are a different use case with return paths to consider, while BIRD158.7 is dealing with on-die buffers. Arpad is concerned with the sentence which defines the ground symbol. Mike thought that this is an issue in BIRD158.7. Michael agreed. Arpad stated that we should do what is technically correct based on Vladimir's DesignCon Summit presentation where he discussed how to handle the referencing of S-parameters in detail. Bob asked, based on that presentation, if the reference can be anything other than node 0, and if all the references in the channel have to be the same. Arpad stated he is discussing this case with Vladimir. He doesn't see a way to connect the Reference to anything other than to node 0 for BIRD158.7. Walter commented that we have to make the LTI assumption for AMI models, and this excludes power aware simulations. Michael suggested to have this question taken up by the ATM group. Arpad will add the BIRD158.7 referencing discussion to the ATM agenda [AR]. Mike asked if the comment regarding consistency with BIRD158.7 can be removed. Arpad asked if BIRD189.5 needs to be changed. Michael replied the likelihood that any changes are needed to BIRD189.5 is low, and the comment can be resolved. Michael stated we have a comment that the 'N' in "N+1" is not defined. He asked if 'N' is defined elsewhere. Bob stated there was a definition later on in the document. Mike noted there is a parenthetical with the definition, and he moved it up to the first use of "N+1". Arpad suggested to do this in other cases where "N+1" is used. Bob noted we should be careful that we are talking about ports and not terminals. Michael suggested to double check that the words "ports" and "terminals" are used correctly. Arpad suggested to change the parenthetical to "(where N is the number of ports in the Touchstone file)". Arpad noted the word "ports" is also used for AMS models in the [External Model] sections of the IBIS specification. Mike thought that we do not need to define 'N' in the same section where it is already defined, but he did add the definition in the Number_of_terminals rules section. Michael noted on page 27, we are creating a circuit with the Unused_port_termination options. We state the value of the resistance to terminate the unused port with but not what the other end of the resistor connects to. Bob suggested to say that it connects to the N+1 reference terminal. Arpad asked if we have that stated elsewhere. Michael noted that we do specify the connection, but it is later on page 27. He suggested to move this definition up to the "Resistance" paragraph. Arpad suggested to add this information to the first sentence of the "Resistance" paragraph. Mike asked if we should be a bit more specific with the names of the Unused_port_termination arguments. Randy made a motion to keep these argument names as they are. Arpad seconded. There were no objections. Bob asked about the "Reference" section. Michael noted we have the same issue with the "Reference" paragraph. Arpad suggested to make the same change to the "Reference" section. Randy agreed. Mike suggested some editorial changes to make the "Reference" and "Resistance" paragraphs to sound the same. Mike to send out BIRD189.5_draft17_v3 [AR]. Next Meeting: The next meeting will be February 21. Mike moved to adjourn. Arpad seconded. The meeting adjourned without objection. Task List BIRD189.5 editorial additions/changes to be completed: