IBIS Macromodel Task Group Meeting date: 10 March 2009 Members (asterisk for those attending): Adge Hawes, IBM Ambrish Varma, Cadence Design Systems Anders Ekholm, Ericsson * Arpad Muranyi, Mentor Graphics Corp. Barry Katz, SiSoft * Bob Ross, Teraspeed Consulting Group Brad Brim, Sigrity Brad Griffin, Cadence Design Systems * David Banas, Xilinx Donald Telian, consultant Doug White, Cisco Systems * Eckhard Lenski, Nokia-Siemens Networks Essaid Bensoudane, ST Microelectronics * Fangyi Rao, Agilent Ganesh Narayanaswamy, ST Micro Gang Kang, Sigrity Hemant Shah, Cadence Design Systems Ian Dodd, Agilent Jerry Chuang, Xilinx Joe Abler, IBM John Angulo, Mentor Graphics John Shields, Mentor Graphics Ken Willis, Cadence Design Systems Kumar Keshavan, Sigrity Lance Wang, Cadence Design Systems Luis Boluna, Cisco Systems * Michael Mirmak, Intel Corp. * Mike LaBonte, Cisco Systems Mike Steinberger, SiSoft * Mustansir Fanaswalla, Xilinx Patrick O'Halloran, Tiburon Design Automation Paul Fernando, NCSU Pavani Jella, TI * Radek Biernacki, Agilent (EESof) * Randy Wolff, Micron Technology Ray Comeau, Cadence Design Systems Richard Mellitz, Intel Richard Ward, Texas Instruments Sam Chitwood, Sigrity Sanjeev Gupta, Agilent Shangli Wu, Cadence Design Systems Sid Singh, Extreme Networks Stephen Scearce, Cisco Systems Steve Pytel, Ansoft * Syed Huq, Cisco Systems Syed Sadeghi, ST Micro Terry Jernberg, Cadence Design Systems * Todd Westerhoff, SiSoft Vladimir Dmitriev-Zdorov Vikas Gupta, Xilinx Vuk Borich, Agilent * Walter Katz, SiSoft Zhen Mu, Cadence Design Systems ------------------------------------------------------------------------ Opens: -------------------------- Call for patent disclosure: - No one declared a patent. ------------- Review of ARs: - Todd: invite Synopsys to join us for a syntax review - TBD - Walter: send updated IBIS Interconnect SPICE proposal to Mike L for posting - Done - Mike L: post updated IBIS Interconnect SPICE proposal - Done - Todd: Start a BIRD for IBIS s-params - TBD - Arpad: Write parameter passing syntax proposal (BIRD draft) for *-AMS models in IBIS that is consistent with the parameter passing syntax of the AMI models - TBD - TBD: Propose a parameter passing syntax for the SPICE - [External ...] also? - TBD - Arpad: Review the documentation (annotation) in the macro libraries. - Deferred until a demand arises or we have nothing else to do ------------- New Discussion: Walter showed TstonefileBird.doc : - A number of issues remain in this document - The statement of the issue invokes broadband analog models - This adds the Tstonefile parameter to [Model] - The expected use is Tstonefile on both sides of the channel - Impulse response for AMI is TXTstone+channel+RXTstone - The TxTstonefile converts an impulse to buffer output - Arpad: So this is separate from the channel model? - Walter: Yes, the package models are in the channel - Walter: IBIS does not model differential capacitance - The Tstonefile can do this - Fangyi: Is RX grounded when we measure? - Walter: Ground for Tstonefiles is a "political" topic - Todd: The question is if it is measured differentially or single ended - Walter: it can be differential or single-ended - Arpad: Will this allow differential coupling? - Todd: Does a Tstone file declare what kind it is? - Walter: You have to read a whole TS1 file to inspect the contents - TS2 has more info - Todd: The intent is for IBIS to point to a TS file and know what to do - Walter: For TS1 files some info has to go into IBIS - This is not specified yet - The IBIS tool has to look into the file to see if it is TS1 or TS2 - Walter: We need to handle more corners than just typ/min/max - Bob: Should we flag if a call is intended for single ended or diff or both? - Arpad: AMI accepts only one or the other - Fangyi: How do we connect this to the GetWave function? - Walter: We don't, this is input to Init, not GetWave - Arpad: The [Model] keyword is inherently single ended - We added a few exceptions - There will be a lot of rules and exceptions around this - Is [Model] not the best place for this new parameter? - Maybe a new keyword is better - Walter: It could be a [Tstonefile] keyword - Todd: There could be another Model_type to allow this - Walter: That limits the uses of Tstonefile - The type could be something like SerDes_RX - If a Tstonefile is there it should be used - You could have an HSPICE model, this is similar - Todd: If there is any problem, a plain B element would be the fallback - Bob: The separate keyword should work well - Walter: This also can solve the corners problem - Todd: [Model] requires parameters that aren't really needed - Like Vinl/Vinh for serdes - Todd: If we have [TStonefile] with no [Model] backup, s-param is required - The only option is a linearized model of the s-param - Bob: We could have [Model] with no data except Tstonefile - It could have phony [Ramp] - Arpad: This is adding baggage to IBIS - Bob: It probably would be read by a new simulator anyway - The [Pin] map is useful "baggage", however - Arpad: We should provide a fallback for simulators that don't handle Tstonefile - Walter: The only simulator without s-param support is Berkeley SPICE - Mike L: How about a Fallback_model parameter for [Tstonefile] keyword? - Todd: The we would have unreferenced [Model]s - Mike L: The new reference is a reference, and should be counted - Arpad: Editorial change, do we need repetition? - Walter: It doesn't hurt to be redundant - Bob: We should restrict this to 2 or 4 ports, not N ports - We don't know how the s-param is generated - Walter: S-params can be combined without simulation - Todd: You derive it any way you can - Arpad: Will these be created for AMI only? - It is not clear how to connect and ground s2p vs s4p - Walter: Ground is meaningless for s-params - Todd: The return path is embedded in the s4p - Do we need the other half of the s4p? - How do we reasonably excite the circuit? - An s4p is replaced with 2 s2ps for single ended mode - Arpad: Is this in series, or a pad to ground shunt? - Walter: It is between an ideal source and the pad - Todd: Why bother with the ideal source at all? - It is equivalent to an AC short - Todd: We want to reuse data the vendors have - They have s4p data - Walter: Kumar used infinite impedances to ground with s2ps - This is what it looks like to the outside world - Arpad: An infinite impedance can be implemented as a current source - Arpad: Will Walter continue working on this? - Walter: No plans for now - Arpad: Hopefully Synopsys can join us next week Next meeting: 17 March 2009 12:00pm PT -----------