IBIS Macromodel Task Group Meeting date: 14 January 2014 Members (asterisk for those attending): Agilent: Fangyi Rao * Radek Biernacki Altera: David Banas Julia Liu Hazlina Ramly ANSYS: Samuel Mertens * Dan Dvorscak * Curtis Clark Steve Pytel Luis Armenta Arrow Electronics: Ian Dodd Cadence Design Systems: Terry Jernberg * Ambrish Varma Feras Al-Hawari * Brad Brim Kumar Keshavan Ken Willis Cavium Networks: Johann Nittmann Celsionix: Kellee Crisafulli Cisco Systems: Ashwin Vasudevan Syed Huq Ericsson: Anders Ekholm IBM: Greg Edlund Intel: Michael Mirmak Maxim Integrated Products: Mahbubul Bari Hassan Rafat Ron Olisar Mentor Graphics: * John Angulo Zhen Mu * Arpad Muranyi Vladimir Dmitriev-Zdorov Micron Technology: * Randy Wolff Justin Butterfield NetLogic Microsystems: Ryan Couts Nokia-Siemens Networks: Eckhard Lenski QLogic Corp. James Zhou Andy Joy SiSoft: * Walter Katz Todd Westerhoff Doug Burns * Mike LaBonte Snowbush IP: Marcus Van Ierssel ST Micro: Syed Sadeghi Teraspeed Consulting Group: Scott McMorrow * Bob Ross TI: Casey Morrison Alfred Chong Vitesse Semiconductor: Eric Sweetman Xilinx: Mustansir Fanaswalla Ray Anderson The meeting was led by Arpad Muranyi ------------------------------------------------------------------------ Opens: - None -------------------------- Call for patent disclosure: - None ------------- Review of ARs: - Walter send presentation to Mike for posting - Done ------------- New Discussion: Package Model proposal from Walter Katz: - Arpad: Is there any more to discuss? - Walter: I will have a presentation at the upcoming summit. - Bob: The title has "IBIS-ISS" but it covers both that and Touchstone. - Both "buf" and "mod" are used, and it is not always clear if they are the same. - Walter: Pin, pad, and buf are separate items. - Where "model" is used it is a model definition name. - "IO" is a model type. - Bob had objected to "+" and "-", preferring "p" and "n". - Bob: On slide 17 "V" should be separated from "IO" with underscore. BIRDS 163, 164, 165 - Arpad: BIRDs 163 and 164 could be merged, are co-dependent. - BIRD 165 is independent of the others. - Arpad showed BIRD 164. - Arpad: This makes [External Circuit] usable for package models. - There was a suggestion that "Package_Model" could have a value. - Bob: It could be [External Package Model]. - Arpad showed BIRD 163. - Arpad: Black text is unchanged, red text is changed. - This adds the ability to connect to [Model] terminals directly. - We found a way to cascade model with external circuits. - Each model is instantiated by [Pin], but the circuit models are aligned to connect. - Bob: [Model] includes [External Model]? - Arpad: Yes. - The first example drawing has changed significantly. - One circuit describes on-die interconnect, another the package. - The numbers after the colons indicate the associated pin numbers. - Radek: Walter's proposal has a buffer concept that is similar. - Walter: Agree, this is the same functionality with a different syntax. - Why can't a die node be anything other than die pad?. - Are die nodes die pads? - Arpad: Yes. - There are buffer nodes that exist only for internal connections. - [Node Declarations] would only be needed for internal die nodes. - The main change for [Circuit Call] is new reserved node names for buffer ports. - Arpad noted that the BIRD has a "questions" section. - Arpad: The previous example Figure 29 showed features not supported by IBIS. - This is corrected in this BIRD. - Power and ground connection would work in concert with new supply bus keywords. - John: This shows everything it can do in one figure. - Ambrish: BIRDs 145 and 125 would be rejected? - Arpad: Yes, once this is approved. - Bob: [External Model] could be called too? - Arpad: No they are called only from [Model]. - [Circuit Call] is only needed once to instantiate the circuit. - Walter: What do power ports connect to where the power comes from multiple pins? - Arpad: The spec is vague on that. - If [Pin Mapping] is used it is more clear. - There has to be one buffer per signal pin. - There could be joins and splits but it has to end at one buffer instance. - John: We have examples of this, although no picture. - Walter: Where does the model in D come from? - Arpad: I haven't worked out how [External Model] would work with this. - John: IC vendors have not often used circuit calls to instantiate buffers. - Mike: Pin names might allow colons, the instantiation syntax might be confusing. - Walter: Aren't the "pos" and "Neg" reserved ports redundant? - Arpad: Those are for series switch models, they are different ports. - Walter: In the figure pads 3a and 3b go to one pin? - Arpad: Yes. - Walter: Where does the output of the driver in C go to? - Arpad: This meets the rule that one pin goes to one model instance. - It was in the previous spec, I didn't delete it. - Walter: We should have examples of this matching some of my examples. - For example, SDRAM, FPGA, and ASIC examples with realistic package models. - We should compare our proposals for functionality. - Bob: Does this deprecate legacy buffers? - Arpad: Nothing is deprecated. - Walter: The package model could be other than IBIS-ISS? - Arpad: Yes. - Bob: I would like to see a legacy buffer added to the figure. - Arpad: That might be confusing. ------------- Next meeting: 21 January 2014 12:00pm PT ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives