From: ibis-interconn-bounce@freelists.org on behalf of Mirmak, Michael Sent: Tuesday, November 27, 2012 9:38 PM To: IBIS-Interconnect (ibis-interconn@freelists.org) Subject: [ibis-interconn] Minutes, Nov. 21 IBIS-Interconnect Task Group and Nov. 28 Agenda ====================================================================== IBIS INTERCONNECT TASK GROUP MEETING http://www.eda.org/ibis/interconnect_wip/ Mailing list: ibis-interconn@freelists.org ====================================================================== Next meeting: Nov. 28, 2012 8 AM US Pacific Daylight Time Agenda: * Attendance * Call for Patents * Agenda and Opens * Decision: Parameter Trees as the Preferred Format for EMD? * Schedule for Work * Next Meetings’ Schedule/Agenda For international numbers, please contact Michael Mirmak. ......................................................................................................................................... Join online meeting https://meet.intel.com/michael.mirmak/QZ193W0C First online meeting? [!OC([1033])!] ......................................................................................................................................... <--- Reservationless Bridge – Do not edit or remove --- 916-356-2663, 8-356-2663, Bridge: 2, Passcode: 8625431 Speed dialer: inteldialer://2,8625431 -----------------------------------------------------------------> Note: in case of issues with Lync we will use the WebEx noted at the bottom of this message ====================================================================== Attendees, Nov. 21 Agilent Technologies Radek Biernacki ANSYS Luis Armenta Cadence Design Systems Brad Brim Intel Michael Mirmak Mentor Graphics Arpad Muranyi Micron Technology Justin Butterfield Signal Integrity Software Walter Katz Teraspeed Consulting Group Bob Ross Minutes No patents were declared. No opens were raised. Michael Mirmak thanked Randy Wolff for his fantastic efforts in running the meeting and taking minutes. The team reviewed the discussions in the previous meetings regarding EBD and/or EMD for package interconnect. Walter Katz noted that EBD is strictly post-layout and pin-oriented with pins on the perimeter of the structure. Walter summarized his proposals from the previous meetings and presentations: the first focus should be to implement EMD; his second presentation was on identifying the ports in a .ibs file, which are more complicated than those in the EBD/EMD. Bob Ross suggested that a discussion of general implications be held first, with a general EMD proposal to follow. Radek Biernacki asked that, as the team was addressing packaging first, whether an EMD effort would distract from that. Arpad Muranyi added that rejecting all analog BIRDs was also proposed; would that be addressed as part of this discussion? Walter made two observations: 1) We need a multi-chip module solution just as badly as a packaging/on-die solution 2) EBD/EMD will deal with multi-chip modules as well as connectors and other arbitrary interconnects. The format is strictly post-layout, with all subcircuits using pins explicitly. This will eliminate problems of die pads and pin pads being different, as there are no die pads defined. Interconnect SPICE circuits are between defined between the pins. "Everything is a connector" (or a cable) in this approach and a general package solution falls out of this solution. Walter noted that he will eventually make a motion to adopt EMD. Bob asked whether EMD can link to another EMD. Walter responded that it can. Michael asked several questions: 1) Does EMD replaces EBD? Would this be a separate document? 2) What is the proposed hierarchy of IBIS, IBIS-ISS, EMD? 3) What is the relationship of EMD to existing specifications? Walter noted several options: 1) Continue calling the format an "EBD" but note that the “B” (board) is more generic now; any substrate, including multi-chip modules and connectors, would be supported, plus extended netlists/model-connection protocol or netlist that the package or IC vendor delivers 2) Same as (1) but call the new specification “EMD” 3) Leverage a trivial amount of EBD (pinlist and component list only), while the main file data will be based on a parameter tree syntax Bob noted that other options should be considered if people suggest them. Walter answered Michael's questions: 1) Interconnect is represented by IBIS-ISS syntax 2) No deprecation of EBD is proposed Bob stated that he would prefer to get rid of fork/endfork syntax and see EMD as a separate specification. The downside of the new format is that the circuit path is unclear. Walter responded that, if you have a path and you trace out that (electrical) path, you have an extended net; we have to add this concept of extended nets to EMD. This is technically inaccurate, but we best describe it as involving low impedance connections (ILN is small) or a DC-blocking path. He added that EMD can be an endpoint and can contain no components at all. Michael asked whether this would be a new document. Walter expressed his preference for a new, separate document but he and Bob noted that it could be a new chapter in the IBIS specification. He would split out EBD from IBIS and would make EMD separate. Bob suggested a fourth option, similar to IBIS but non-tree based, as the IBIS-ISS interface is nodal. Radek asked what is the advantage/disadvantage of having the document in the tree format for EMD. Walter responded that one branch of the EMD file can contain packaging data (one parser can deal with both) and that the package data can be a branch within that parameter tree. Radek asked whether, if we start with EBD, whether we get rid of any parts of it. Bob suggested that Fork/Endfork would be removed, along with RLC descriptions without coupling; weakness of EBD vs. EMD is its coupling treatment. Walter restated his proposal options: 1) EBD with “B” being more generic 2) Same as (2) but naming change to EMD 3) EMD with parameter tree as a separate document; reuse a trivial amount of EBD concepts 4) New document but not parameter tree Walter added that strongly wants parameter tree as new document. Bob stated that he does not want to see changes to or deprecation of EBD. Walter agreed - no deprecation, but another, separate document. Bob suggested focusing work on the parameter tree format, with specification development for EMD considered separately. The next meeting will be Nov. 28. Michael proposed that the parameter tree format be the focus of discussion, with a decision to be reached by the end of the meeting. ====================================================================== In case of Lync issues only, we will switch to WebEx as noted below. Meeting Number: 732 940 715 Meeting Password: IBIS ------------------------------------------------------- To join this meeting (Now from mobile devices!) ------------------------------------------------------- 1. Go to https://sisoft.webex.com/sisoft/j.php?J=732940715&PW=NNWY2NmRmZTY0 2. If requested, enter your name and email address. 3. If a password is required, enter the meeting password: IBIS 4. Click "Join". 5. Follow the instructions that appear on your screen. ------------------------------------------------------- Audio conference information ------------------------------------------------------- Call-in toll number (US/Canada): 1-650-479-3208 Access code:732 940 715 http://www.webex.com IMPORTANT NOTICE: This WebEx service includes a feature that allows audio and any documents and other materials exchanged or viewed during the session to be recorded. By joining this session, you automatically consent to such recordings. If you do not consent to the recording, discuss your concerns with the meeting host prior to the start of the recording or do not join the session. Please note that any such recordings may be subject to discovery in the event of litigation.