================================================================================ IBIS INTERCONNECT TASK GROUP Mailing list: ibis-interconnect@freelists.org ================================================================================ Attendees from March 8, 2023 Meeting (* means attended at least using audio) ANSYS Curtis Clark Intel Corp. Michael Mirmak* Michael Brownell Keysight Technologies Ming Yan Marvell Steve Parker MathWorks Walter Katz* Micron Technology Justin Butterfield* Randy Wolff* Siemens EDA Arpad Muranyi* ST Microelectronics Aurora Sanna Teraspeed Labs Bob Ross* University of Illinois Jose Schutt-Aine Zuken USA Lance Wang* 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 1, 2023 meeting. Michael displayed the minutes. Arpad Muranyi moved to approve the minutes. Randy Wolff seconded. The minutes were approved without objection. Review of ARs: - Michael to check if the Network Analyzer manuals documenting the original Touchstone can be posted to the IBIS website [AR]. - Michael suggested to keep this open, as he has not heard back from his Keysight contacts yet. - Everyone to test the proposed duplicate option line syntax in their internal parsers. - Michael noted we will discuss this today. Opens: - None. Discussion: Touchstone X Discussion: Michael asked if anyone would like to share about their internal parser behavior. Michael commented the official Touchstone parser gives a warning for the additional option lines. The proposal is to leverage existing parsers in the industry and add new features in Touchstone without causing issues in the existing parsers. Michael asked if there was a specific proposal Walter Katz was making to have the Touchstone 2 keywords as optional. Walter replied he did not address this in his slides. His preference would be to not have keywords by themselves, but we could have the keywords preceded by the pound symbol. Bob Ross was concerned that the Touchstone 2 parser source code has already been purchased by several companies, and we still need to support the Touchstone 2 specification. Walter suggested that Touchstone 2 will just not be improved further, and he would propose a new direction to use the additional option line with the features that users want. Michael stated he found several tools that read and write Touchstone 2. The problem is that he has not seen Touchstone 2 files in use. Walter agreed with this. Randy noted they have used Touchstone 2 for package models to reduce the file size. They do occasionally use the per port reference impedance feature. Arpad noted he has seen models that violate the Touchstone 1 specification that have per port impedance values on the option line. Randy noted he has seen this as well with multiple values for the reference impedance on the option line. Michael asked how this would work in the parser. Randy noted these models will fail in the official Touchstone parser. Lance Wang noted his tool will not fail on additional option lines. Arpad noted he also tested this in his internal parser, and it is okay to have an additional option lines in his tool. Michael will ask the IEEE 370 group what is their preference for the direction and what are the features they would like to see [AR]. Bob noted that we know they support port mapping with comments. Bob asked if they use Touchstone 2. Michael will find out. Bob was concerned we would be competing with the IEEE 370 proposals. Arpad noted the advantage with Walter's proposal is that old Touchstone 1 parsers will not die on the new Touchstone files. He did not personally like the change in direction of the specification. Arpad asked if it is the keyword style or the features in Touchstone 2 that people don't like. Walter noted there were some features, such as the Matrix format, which break existing parsers. And, you cannot always down convert to Touchstone 1. Walter suggested that a factor of 2 is not significant enough of a file size reduction, and he thought that the sparse format can provide more significant file size reduction. Arpad reiterated his question about what people don't like about Touchstone 2, and he noted it is difficult to move forward without answering this question. Walter replied Touchstone 2 added new requirements and changed to the keyword structure. Arpad asked what is the value of having new features that require the data to be recalculated. He also suggested that the keyword structure is more readable. Walter noted, for the impedance per port feature, if the user does not simulate the ports with a different reference impedance, the model can still work. Arpad commented that this depends heavily on how the end user uses the model in the simulations which cannot be guaranteed. Walter stated that the user would have to use a tool which supports that feature completely. Arpad asked if there are tools which do not support Touchstone 2 files. Randy replied that he has had to down convert Touchstone 2 to Touchstone 1 for customers. Bob commented that there have been 9 purchases of the Touchstone 2 parser. Arpad would like to understand the benefit and need for using the duplicate option line syntax. For many new features, the data will have to be recalculated. Michael asked if there is any discussion about Touchstone around the SPIM proposal in the IBIS-ATM task group. Arpad replied that there is an assumption in the SPIM discussion that Touchstone has the features to support it. Michael asked if it is worthwhile to ask about Touchstone 2 usage in ATM and if Touchstone 2 is required for SPIM. Arpad will send an email to the ATM reflector to ask about Touchstone 2 [AR]. Michael asked, if there is not a must have feature in Touchstone 2 and the format is not supported in existing parsers, could we have a combination of the proposals. We could make the Touchstone 2 keywords option lines. But, as soon as you change the network data, the parser needs to be updated. Arpad moved to adjourn. Bob seconded. Next Meeting: The next meeting will be March 15, 2023. ================================================================================ Bin List: 1. Draft Touchstone document separating version 1.0 and 2.0 2. Encapsulating Touchstone 1.0 data 3. Touchstone 2.0 draft with TSIRD3 and TSIRD4 4. Pole-residue format 5. Port naming 6. Alternatives to the Touchstone 1.0 option line 7. Enable Cascading of S-parameters Through W-element 8. Touchstone 3.0 draft outline (dependent on several items above) Tabled ARs: - Arpad to give an example of the physical connectivity needed for EMD automation.