IBIS Macromodel Task Group Meeting date: 11 July 2023 Members (asterisk for those attending): Achronix Semiconductor: Hansel Dsilva Amazon: John Yan ANSYS: * Curtis Clark * Wei-hsing Huang Aurora System: Dian Yang Cadence Design Systems: * Ambrish Varma * Jared James Google: Hanfeng Wang GaWon Kim Intel: Michael Mirmak * Kinger Cai Chi-te Chen * Liwei Zhao Keysight Technologies: Fangyi Rao Majid Ahadi Dolatsara Stephen Slater Ming Yan Rui Yang Marvell: Steve Parker Mathworks (SiSoft): Walter Katz Graham Kus Micron Technology: Justin Butterfield Missouri S&T: Chulsoon Hwang Yifan Ding Zhiping Yang Rivos: Yansheng Wang SAE ITC: Michael McNair Siemens EDA (Mentor): * Arpad Muranyi * Randy Wolff Teraspeed Labs: * Bob Ross Zuken USA: Lance Wang The meeting was led by Arpad Muranyi. Curtis Clark took the minutes. -------------------------------------------------------------------------------- Opens: - Arpad noted that he would be traveling on July 18th and unable to chair the meeting. He asked whether we should cancel the meeting or have someone else run it. Kinger suggested that we cancel the meeting. He said he would continue to incorporate feedback from Fangyi on the PSIJ proposal, and we could discuss that in two weeks. No one objected. The meeting scheduled for July 18th will not be held. ------------- Review of ARs: Kinger: Work offline with Fangyi to address Fangyi's suggestions on the PSIJ Sensitivity draft11. - Done. Kinger had met with Fangyi and sent out draft12. -------------------------- Call for patent disclosure: - None. ------------------------- Review of Meeting Minutes: Arpad asked for any comments or corrections to the minutes of the June 27th meeting. Curtis moved to approve the minutes. Kinger seconded the motion. There were no objections. -------------- New Discussion: New BIRD Draft related to ibischk parser BUG244: Arpad presented a new BIRD draft aimed at addressing one aspect of BUG244. The proposal contains a small addition to the EMD [Designator Pin List] keyword's Usage Rules, which explicitly states what is currently (IBIS 7.2) an implicit requirement. IBIS 7.2 states that all pins with the same bus_label must have the same signal name. It was implicit that this requirement also applies to all [Component]s included in the EMD, so this new text is added: within EMD files as well as within any IBIS [Component] referenced in [EMD Parts] If the EMD file uses an included IBIS file that is pre-IBIS 7.0, the ibischk7.2 parser currently issues an error stating: E5810 - The file .ibs has [IBIS Ver] <3.2>; however, IBIS 7.0 compatibility is required for IBIS files referenced by EMD However, this is overly strict. Included IBIS files need not be version 7.0 or higher. They only need to meet IBIS 7.0 bus_label requirements. Randy said that additional changes to address BUG244 will include modifying the E5810 message to indicate the bus_label compatibility requirement and only displaying E5810 if the IBIS 7.0 bus_label requirements are violated by the pre 7.0 IBIS file. Bob said Arpad's proposal was a simple change to correctly state things in the specification. Bob moved that Arpad submit the proposal to the Open Forum as an official BIRD. Randy seconded. There were no objections. Arpad took an AR to send the proposal to Lance Wang (IBIS Chair) to post as an official BIRD. (Done, BIRD 225). PSIJ Sensitivity Discussion: Kinger shared draft12 of the proposal, which he had emailed to the ATM list. He said that it incorporated some feedback from Fangyi, but not all of the feedback had been addressed in this version. Changes in draft 12: Requirements Section: - In the [PSIJ Sensitivity Signal] description, it now explicitly states that [PSIJ Sensitivity Signal] is a superset of [ISSO PU] and [ISSO PD] and includes their contributions. Introduction Section: - Tjeye is defined as the total jitter of the eye diagram resulting from SI and PI cosimulation. - Kinger noted that Fangyi had suggested that the paragraphs describing equations 1 through 3 might not be necessary in the BIRD (specification) itself. Arpad suggested that Kinger could consider folding those paragraphs into the Definition or Background sections. - The description of figure 2 explicitly mentions that the iFFT of the jitter spectrum is the jitter waveform in time domain. - Equation 7: The denominator is 2*A (it was previously A). Kinger observed that Fangyi had also suggested the equation 6, which defines A, was not necessary: A = (Vmax - Vmin)/2 However, Kinger said he wanted to keep equation 6 to explicitly define A in terms of Vmax and Vmin, which represent the worst-case deviations from Vtypical. - Equation 8: This equation is new and explicitly defines the Jitter transfer function of each CKT_i in terms of its input signal Sin and its output signal Sout. - New equations 9, 10-1, 10-2: These equations explicitly define the jitter output of each CKT as the jitter of its input signal times its jitter transfer function (JTF) plus the additional jitter it introduces. The assumption, which is captured in 10-1, is that there is no jitter on the initial signal input to the first CKT. - A new paragraph below figure 5 explicitly defines the effective measurement locations of PSIJ_Vcc_m_N for the Tx (output pin of the last stage with a typical test load) and Rx (input to the latching circuit). Kinger said that it's typically the IP vendor's responsibility to provide the final cumulative PSIJ Sensitivity for the Tx and Rx. He said that perhaps the example provided in the BIRD is unnecessary. Arpad asked how the IP vendor would know what to provide in the model if we took the example out of the actual specification changes and put it in another section of the BIRD. Arpad suggested that we just make it clear that it is an illustrative example and not necessarily the way it must be done. Kinger noted that the introductory sentence (line 104) states that it is an illustrative example. - PSIJ Sensitivity Rail Description: Kinger noted that Fangyi had suggested explicitly stating that the two signal names are for Power and GND respectively. Kinger said he would continue incorporating feedback from Fangyi and today's meeting and send out draft13. - Ambrish: Motion to adjourn. - Kinger: Second. - Arpad: Thank you all for joining. New ARs: Kinger: Send out draft13 of the PSIJ Sensitivity proposal including more feedback from Fangyi and the ATM meeting. ------------- Next meeting: 25 July 2023 12:00pm PT ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives