IBIS Macromodel Task Group Meeting date: 09 Oct 2012 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 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 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 Sigrity: Brad Brim Kumar Keshavan Ken Willis 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: We will need to discuss meeting dates during the summits -------------------------- Call for patent disclosure: - David asked about the purpose of this requirement. - Michael M: It is an IP protection requirement from TechAmerica - We would have to know what the terms are for use of patentable ideas - The TechAmerica operating manual has specifics - David: Is it required when the patent is initiated or granted? - Michael M: I will find out - Walter: IEEE requires this to know if something they are discussing might be patented later - There was a stateye development that was patented later, causing turmoil - David: Then the announcement must be before the patent is granted - Walter: Yes, for example I could patent dependency tables and that would impact our spec - Michael M. read the official TA language, found at http://fileshar.es/Mx32Rzi ------------- Review of ARs: - Walter post BIRD 123.4 draft 4 with updates - Done - Arpad: A draft 5 was also sent - Walter post BIRD 121.2 draft 5 with updates - Done ------------- New Discussion: Interconnect task group report: - Michael M: The minutes were sent recently - The question is update, deprecate or fork? - The consensus was not to have a new specification BIRD 121: - Arpad showed BIRD 121.2 draft 5 - Arpad: I hope we can vote on this today - A question came up about slashes - I suggest requiring the ending slash - Walter: That should be optional - Arpad: Why does it now say there should be no slash? - Walter: My tool works that way, but it could be optional - David: "D:" is not a relative path - John: That goes to whatever is the CWD on drive D, it is relative - Walter: Relative paths help especially when running on simulation farms - David: Any absolute paths in zip files are likely to fail - Arpad: Why would the tool need to use relative paths when it knows the full path to pass? - John: DLLs for different OSes each need a valid path, and relative paths help? - Walter: Yes, it may be different on each OS - Arpad: Does the spec have to discuss what the EDA tool does internally? - Walter: The tool only needs to know where the files are delivered - James: It should be possible to find out where the DLL file itself is - John: We might want to set the DLL CWD on a per simulation basis - Mike L: DLLs have no process, no CWD, no ability to call functions in the host program - Walter: The DLL still needs to find it's supporting files - Ambrish: Maybe the name should be Supporting_Files_Path - Arpad: By knowing the DLL location we can find the other files - Arpad: The CWD might change by the time the DLL does it's work - Walter: The DLL itself might even change the CWD - John: We have no loss of flexibility by having relative paths - The DLL does rely on the calling tool for CWD - It should rely on DLL_Path instead - Arpad: We may need to reword the sentence somewhat - Walter: We could do that offline - I still object to requiring a slash - It is very easy for software to just add missing slashes - It would be more consistent to not allow it - James: This should not be a problem - Arpad: We will not vote on this now Arpad: I checked BIRD 150 and found no technical losses in the new format version. BIRD 123: - Walter showed BIRD 123.4. draft 5 - Walter explained the changes - David: Why is there a "2" in front of Dj but not the others? - Walter: Not sure. Dj is half peak to peak. - David: Is Rand() defined clearly? - Walter found the definition - Bob: It should appear before its first use - Walter: rand() is defined where first used, and so is gaussian_rand(). - Arpad: We should vote on this next week - Walter: This will be posted as draft 6 with changes accepted - Arpad: We should not do that - Was a draft 6 already sent? - Walter: I don't have that Walter: There was an interesting email from Ambrish about BIRD 150 - Walter showed the email - Walter: I have no objection to changing the examples - Ambrish: It is not specified if the conditions are OR or AND - Walter: They are all AND except for PWL - Only one of those would be allowed - Ambrish: Why is Corner not defined? - Walter: That is an intrinsic parameter - Arpad: Corner is well defined in IBIS 5.1 - Ambrish: Corner is a format, not a parameter - Why do we need these tables? - Walter: Adge has asked for it - For example he needs to know which touchstone file to use - Arpad pointed out some typographical errors - Walter and Ambrish can clear up the language offline Ambrish: Please look at the latest backchannel BIRD that has been posted - Walter: It would help to have an operational backchannel DLL to look at this - Ambrish: I have those, but may not be able to post them ------------- Next meeting: 16 Oct 2012 12:00pm PT Next agenda: 1) Task list item discussions ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives