IBIS Macromodel Task Group Meeting date:17 August 2021 Members (asterisk for those attending): Achronix Semiconductor: Hansel Dsilva Amazon: John Yan ANSYS: * Curtis Clark * Wei-hsing Huang Cadence Design Systems: * Ambrish Varma Ken Willis Jared James Google: Zhiping Yang Intel: Michael Mirmak Kinger Cai Alaeddin Aydiner Keysight Technologies: * Fangyi Rao Radek Biernacki Ming Yan Todd Bermensolo * Rui Yang Luminous Computing David Banas Marvell Steve Parker Mathworks (SiSoft): * Walter Katz Mike LaBonte Micron Technology: * Randy Wolff * Justin Butterfield Missouri S&T Chulsoon Hwang Siemens EDA (Mentor): * Arpad Muranyi Teraspeed Labs: * Bob Ross Zuken USA: * Lance Wang The meeting was led by Arpad Muranyi. Curtis Clark took the minutes. -------------------------------------------------------------------------------- Opens: - Arpad said the Editorial task group wanted our input on the terminology to use for the first Tx and the last Rx in a channel. He said the decision will likely affect what we do with BIRD211.3 as well. ------------- Review of ARs: - Arpad noted an implicit AR for Fangyi to send BIRD211.3 draft5 to ATM for people to review. - Done. -------------------------- Call for patent disclosure: - None. ------------------------- Review of Meeting Minutes: Arpad asked for any comments or corrections to the minutes of the August 10th meeting. Walter moved to approve the minutes. Ambrish seconded the motion. There were no objections. ------------- New Discussion: BIRD211.3 draft5: Arpad opened discussion with the Editorial task group's question about the terminology to use for the first Tx and last Rx in a channel. He said it would also affect the language of this BIRD. Arpad summarized some of the possible terminology that had been used: - Upstream and Downstream - Arpad said he didn't like these terms because they were relative to a given location. - Terminal - Arpad said he wasn't sure he liked this choice because terminal is an overloaded word (e.g., as a noun it might be a synonym for node) and could cause confusion. - Primary - Arpad said this also bothered him because primary has other connotations such as "most important". Bob asked what was currently in BIRD211.3 draft 5. Arpad searched: - 1 instance of "primary Tx" - 5 instances of "Terminal Tx" and 6 instances of "Terminal Rx", only two of which were in the text that would actually go into the specification. Randy noted that figure 41 (likely to be figure 43 in IBIS 7.1) would be a good place to add the new terminology we decided to use. Curtis noted that Walter had originally used "initial Tx" for the first Tx and "terminal Rx" for the final Rx. Fangyi had later used terminal for both. Curtis said he had liked Walter's terminology, but he had no issue with Fangyi's symmetric approach. Ambrish and Fangyi both said they liked using "terminal" for both the first Tx and the last Rx. Arpad asked if anyone was opposed. There were no objections. Bob said this question had arisen in the Editorial task group with respect to BIRD201.1, and the terminology would be applied in that BIRD, too. Arpad took an AR to report back to the Editorial task group on the decision to use "terminal". Fangyi then shared BIRD211.3 draft 5. He changed the one instance of "primary Tx" to "terminal Tx". Randy said figure 41 is where we could apply the "terminal" terminology to the first Tx and last Rx. Randy took an AR to review with the Editorial task group whether this same figure should be modified in IBIS 7.1 to add the "terminal" text. Fangyi reviewed the two main changes he had made to this draft in response to Ambrish's comments from the last meeting: 1. In the normal flow section, removed the "Under certain circumstances" paragraph after step 3. Rewrote step 6d and added some of the information from the deleted paragraph as bullets under 6d. 2. Changed the Retimer flows. They are now identical to what was in IBIS 7.0, except that statistical and time-domain flows are broken out separately. Ambrish thanked Fangyi for these changes and said he thought they made the BIRD more readable and consistent with IBIS 7.0. The group continued reviewing Randy's comments and suggestions from the previous draft. Randy noted that in step 6d of the normal flow, item 1 (deconvolution) was written as a paragraph while item 2 (unit impulse response) was written as a list. Ambrish suggested that the list items were really separate notes, and Fangyi added "Notes:" prior to the list in item two and rewrote item 1 in the same fashion to use "Notes:" and a list. Randy noted that there were still multiple instances of double possessive language that we should remove. For example: Rx1's executable model's AMI_Init function... The group decided to remove "executable model's" from all instances. For example, resulting in: Rx1's AMI_Init function... Randy asked that the large figure showing the Redriver statistical flows for the four different values of Tx_Impulse_Input be cleaned up (alignment of text, channel figures, etc.) Fangyi added a note to do this. On page 11, item 5a, Ambrish noted that IBIS 7.0 used the additional phrase "is presented to" when talking about passing the impulse_matrix to the AMI_Init function. Arpad suggested breaking one sentence into two and adding this phrase to the second sentence in the form: "The impulse_matrix is presented to Tx2's AMI_Init, and Tx2's AMI_Init is executed." Fangyi agreed to make similar language changes throughout. Arpad suggested one more final review at next week's meeting, and then hopefully we can submit the new version to the Open Forum. Randy said that if we get it done at the next ATM meeting we can introduce BIRD211.3 at the Open Forum meeting on August 27th. - Ambrish: Motion to adjourn. - Curtis: Second. - Arpad: Thank you all for joining. AR: Arpad to report back to the Editorial task group on ATM's recommendation to use "terminal" to describe the first Tx and last Rx. AR: Randy to check with the Editorial task group to see if Figure 41, the repeater link physical layout, should have "terminal" language added in the upcoming IBIS 7.1. AR: Fangyi to send BIRD211.3 draft 6 to ATM for review. ------------- Next meeting: 24 August 2021 12:00pm PT ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives