IBIS Macromodel Task Group Meeting date: 21 December 2010 Members (asterisk for those attending): Agilent: * Fangyi Rao Radek Biernacki Ansoft: Chris Herrick Danil Kirsanov Ansys: * Samuel Mertens * Dan Dvorjak Deepak Ramaswamy Jianhua Gu Cadence Design Systems: Terry Jernberg Ambrish Varma Celsionix: Kellee Crisafulli Cisco Systems: * Mike LaBonte Stephen Scearce Ashwin Vasudevan Ericsson: Anders Ekholm IBM: Greg Edlund Intel: * Michael Mirmak LSI Logic: Wenyi Jin Mentor Graphics: * John Angulo Vladimir Dmitriev-Zdorov Zhen Mu * Arpad Muranyi Micron Technology: Randy Wolff Nokia-Siemens Networks: * Eckhard Lenski Sigrity: Brad Brim Kumar Keshavan Ken Willis SiSoft: * Walter Katz Mike Steinberger * Todd Westerhoff ST Micro: Syed Sadeghi Teraspeed Consulting Group: * Scott McMorrow * Bob Ross TI: Casey Morrison Alfred Chong Vitesse Semiconductor: Eric Sweetman Xilinx: Mustansir Fanaswalla The meeting was lead by Arpad Muranyi ------------------------------------------------------------------------ Opens: - Arpad said the next meeting would be Jan 4, 2011 -------------------------- Call for patent disclosure: - none ------------- Review of ARs: - Walter submit BIRD to open forum - Done - Arpad update and distribute Defaults BIRD - Done - BIRD 126 now - Walter send example SPICE deck to Mike for posting - Done ------------- New Discussion: Arpad showed the Typos BIRD - Arpad read passages from the BIRD related to input and output strings - Page 5: - Bob: Is the Usage passed to the DLL? - Is it prohibited? - Walter: It should be prohibited - Mike: List based formats like this allow for extraneous content - Programs pick what they want, ignore all else - Bob: We may be using "leaf" ambiguously here - Arpad: All leaves have to begin with one of these keywords - Arpad: Are we ready to vote on this? - Bob: We should vote in January Arpad showed BIRDs 123 & 124: - Walter: Bird 123 is the "jitter" BIRD - We did not understand this well 3 years ago - This redefines how jitter is described to the EDA tool - Only Tx_DCD is reused, with clarification - DCD is important - a 6GHz signal often has a 3GHz clock - Clock DCD is introduced - Pulse width distortion is a data artifact - Rj is just random, Gaussian - Might be caused by using multiple internal data paths of different lengths - Sj is sinusoidal - Might be power supply fluctuation - Arpad: The Usage of these is Info - What is the meaning of how this should be used? - Walter: - The tool could use it for statistical processing - Jitter may come form the model itself - In that case it would be Out - Arpad: How can it influence the jitter - Walter: The Out is from the Init call - Arpad: That should be in the text - Walter: Agree - Bob: Do we disallow Corner for Out parameters? - Walter: That should not be allowed - Walter: Note there is no Dj here - That is pattern dependent - It would have to be handled by GetWave - Michael M: This means the tool is doing most jitter calculation - The model author does not control the calculation - Walter: FUJ is Founded Uncorrelated Jitter - Did not add a parameter for this - Could using existing Tx_Jitter keyword with tables - Fangyi: This sounds related to Rj and Dj - Walter: Rx is very different - Rx has 2 clocks: system and recovered - The recovered clock interacts with both data and the system clock - An Init-only model can use only parameters for clock recovery - There is no Rx_Clock_Recovery_Sj frequency - If no Tx clock ticks the params are used - Otherwise params not used - Scott: A flow diagram would help here - All tools must give the same results - Walter: Agree - Walter: Jitter is in units of time - Rx_Noise is amplitude - Fangyi: If Rx GetWave returns clock ticks does EDA tool use these parameters? - Walter: Yes it could, the jitter may be on the system clock - This is in addition to the jitter from Rx GetWave - Scott: GetWave can use these as inputs - It will impact clock ticks coming out - There are 3 choices for jitter: - 1 Completely model it inside Rx - 2 GetWave perturbs the signal - 3 Pass parameters to the EDA tool to handle it - Walter: It could be Model_Specific too - Arpad: Is there any ambiguity here? - Walter: We need to look at that carefully - Bob: Tx DCD is the only one that allows Corner - Should it also have List? - Walter: It should, and the others should have the same rules - List is not that logical, but it could be allowed - Value, Range, and Corner should do it - The BIRD is mostly consistent with 5.0 BIRD 124: - Walter: Dependency format defined - Tx_Strength_Table is used as an example - This provides "knobs" for users to control - It helps to have setting one parameter change others too - Would have dependent and independent parameters Arpad thanked everyone for a year of good work on AMI ------------- Next meeting: 4 Jan 2011 12:00pm PT Next agenda: 1) BIRD 121-124 discussions ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives