IBIS Macromodel Task Group Meeting date: 17 December 2013 Members (asterisk for those attending): Agilent: Fangyi Rao * Radek Biernacki Altera: David Banas Julia Liu Hazlina Ramly Andrew Joy Consulting: Andy Joy 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 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: - Arpad: I am on the road, may not be very interactive. - Arpad: Reminder that our next meeting will be Jan 7 -------------------------- Call for patent disclosure: - None ------------- Review of ARs: - Walter send updated Package Modeling slides to Mike for posting - Done ------------- New Discussion: Package Modeling: - Walter showed a presentation IBIS-ISS Package Syntax - slide 3: - Walter: I made some changes to Yes/No/Maybe here - slide 4: - The intent is to create a syntax that meets all of our requirements - Slide 6: - Walter: New syntax is in red - Slide 7: - Walter: The Language keyword allows for other future languages - Radek: Does [Package Model] replace the existing keyword? - Walter changed it to [Begin Package Model] - Slide 8: - Bob: The package model name could be a pin number? - Walter: No it is just a label - Walter: The File keyword allows for multiple file names - These might be a file for each corner - Slide 9: - Walter: The Subckt keyword has a Format word Corner or Value - Slide 10: - Walter: The Parameter keyword borrows from AMI but no parentheses - Slide 11: - Walter: Other Formats are possible too. - Each of these should be required to have a typical value. - Randy: We might need dependency tables. - Brad: How many IC vendors will supply this level of complexity? - Will they use Gaussian, Range, and PDF? - Walter: Intel has information about dielectric constant distribution. for example. - Models can be produced statistically. - Brad: This can get complicated. - Intel will not give this to many people. - Slide 12: - Walter added Pin_Mod+, Pin_Mod- for diff pairs - Slide 11: - Walter: Brad asked earlier how much this is needed in the industry - MM: Typ/Min/Max doesn't give the granularity needed for statistical analysis - Arpad: How common will these models be? - MM: Hard to say, we will both create and use the models - Arpad: Will they become public? - MM: Yes, they will go to our customers. - That is not the same as "public" - Brad: Packages are derived using statistical methods, but does that need to be released? - MM: Having a model of every pin is inconvenient - But we need to model the variations over the pins somehow. - Walter: You may have to supply a tool, unless this becomes part of IBIS - Slide 15: - Radek: It is very important to address this issue of using part of a package model - Walter: Agree - Slide 17: - Walter: I have a file from a large company with 30 character pin names - We should allow up to 40 characters - This would require longer line limits - Slide 16: - Randy: I may want an on-die decoupling model plus package model - Walter: Three circuits, pin2buf, power pin2pad, pad2pin? - That would be OK. - You can't couple between signal and power though. - Brad: I would prefer to list all pads if one is listed - A fourth short circuit model could be added to make connections - Randy: For signals the pad really is the buffer - Brad: It would be best not to have groups of subckts that go between different sets of things - We should not split up power and signals so much - Just put all of the pads there - Randy: I probably would - Bob: We would need all pins too - The syntax supports the possibility of having only pads and buffers - There should be a rule - Walter: It might describe a die with no package - Randy: Having the nodes is important for stacked die models - Walter: There can be a flag to say this is a die - Radek: The pins and buffers must all be there because the naming convention implies them - Bob: A buffer can be used with multiple pins - Walter: We ruled that out - Bob: I mean a buffer model definition can be reused ------------- Next meeting: 7 January 2013 12:00pm PT ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives