================================================================================ 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 March 7, 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 March 7 meeting. Mike LaBonte moved to approve the minutes. Bob Ross seconded. The minutes were approved without objection. Review of ARs: - Walter Katz to supply pseudocode for the parser development. - Mike noted that Walter did send this out. - Mike to find missing meeting notes. - Mike noted he did find and post three missing meeting notes. - m081915-interconn.txt - m110317-interconn.txt - m120617-interconn.txt Opens: - None. Summary of ATM: - Arpad Muranyi noted that the replacement draft for BIRD158 was discussed. Walter also gave a proposal on how to fix ground in IBIS and BIRD189. Although, we did not discuss BIRD189 specifics. Michael asked if there were any conclusions from the BIRD158 discussion that would influence BIRD189. Arpad replied that there were not. BIRD189.5_draft18_v3 Discussion: Michael asked if we have agreed on the rail rule relaxation. Arpad noted that Bob made the changes to BIRD189.BIRD189.5_draft18_v3, but that these were not agreed to. Arpad also noted we made a change to BIRD189 that A_gnd is a local ground and not a global ground. We should formally decide that. Michael asked what is meant by the word "local". Arpad noted that local is to the Interconnect Model. Bob noted that he does not want to extend A_gnd to anything other than ground. Per Walter's email where he clarified voltage measurements, the local reference could be connected by EDA tools to the global reference. Bob commented that this becomes a terminology issue. Mike noted that Walter's point was that we should not use the word "ground" but rather call it the reference node. He thought BIRD189 could use some clean up regarding these terms. Michael asked if we have a consensus that BIRD189 should define A_gnd as a local reference, and what is the impact of that change to the text of the document. Arpad stated that we are already doing this for the Touchstone N+1 approach. Mike noted the definition of A_gnd does mention ground. Michael noted that the introductory material has a mention of A_gnd, and this is different from the definition later in the BIRD that will go into the IBIS specification. Arpad stated that the sentence in the introductory text should indicate A_gnd can be connected to the global ground. Bob commented that in the Multi-lingual section, A_gnd was introduced as ground, and we are changing the meaning. He expressed concern that we are getting too far outside of the original intent of A_gnd. He commented that some tools may have a selection mechanism to select what A_gnd connects to. Michael agreed that the EDA tool may give the user a choice how to connect A_gnd. Michael asked if we can remove the word "ground" and replace it with "reference" globally in BIRD189. Bob asked if this would include the statement in the introduction. Michael replied, yes, but we would have to do the replacement intelligently. Mike suggested to look at how this would work in the text. For instance, node 0 would change to global reference. Arpad noted he has concern with the A_gnd sentence in the introduction sentence. We need to be clear on what A_gnd means and state what A_gnd might be connected to. Michael suggested to change the sentence to add the word "allow". Michael asked if the concern was about having the node 0 inside the subcircuit. Mike thought that we cannot disallow this, but we can give warnings. Bob thought it is extreme to not mention ground and always call it a local reference. Mike suggested we should have a more concise definition for A_gnd. We do not have to use the word "ground", as its meaning is not clear. He thought simply stating that A_gnd could connect to node 0 should be sufficient. In the page 94 changes, Mike noted that it says [Component]. Arpad had some concerns with this and its relation to his changes to BIRD158. Bob asked why we are changing this statement and stated that perhaps we should not change the Multi-lingual section. Michael stated that we seem to have some consensus to change "ground" to "reference". Although, we may need to retain the word "ground" in a few places. Bob asked if Multi-lingual section is broken if we do not have this new A_gnd definition. Michael suggested to move the text to the interconnect section. Arpad noted that we already have A_gnd defined on page 7. Bob suggested to add A_gnd as option for Terminal_type to each of the three interfaces and remove ground as an interface on page 7. He also suggested to add a definition of A_gnd in this section and note that it can be on any interface. Michael added the A_gnd definition to page 7. Bob thought that people might object to the word "ground". Michael asked if it is okay to mention ground in this section only. Bob stated he is okay with this approach. Mike asked about the rail rule relaxation that is included in BIRD189.5_draft18_v3. He noted the rail relaxation rules were not agreed to. Michael stated that he could make the A_gnd changes to BIRD189.5_draft18_v2 and BIRD189.5_draft18_v3 would become a fork. Bob suggested to delete the mention of [Component] on page 5. Arpad thought that the term "local" needs to have a scoping defining if it is local to the [Model] or the [Component]. He stated the locality of BIRD189 is under the [Component]. Michael asked if we need to take the Multi-lingual section as a separate issue and possibly a separate BIRD. Arpad stated that we have used A_gnd, and we need to be consistent. Bob noted that we are changing the meaning of the A_gnd as the universal ground. Arpad asked if we should change the name to A_ref and define that as the local reference. Michael thought this pushes the problem out to be resolved later. Bob suggested to make BIRD189.5_draft18_v4 based on BIRD189.5_draft18_v3, and we can take out the rail rule changes if need be. Michael will create BIRD189.5_draft18_v4 with the proposed changes to replace "ground" with "reference" [AR]. Next Meeting: The next meeting will be March 16. Mike moved to adjourn. Arpad seconded. The meeting adjourned without objection. Task List BIRD189.5 editorial additions/changes to be completed: