IBIS Macromodel Task Group Meeting date: 26 August 2014 Members (asterisk for those attending): Altera: David Banas ANSYS: Dan Dvorscak Curtis Clark Avago (LSI) Xingdong Dai Cadence Design Systems: Ambrish Varma Brad Brim Kumar Keshavan Ken Willis Ericsson: Anders Ekholm Intel: Michael Mirmak Keysight Technologies: * Fangyi Rao * Radek Biernacki Maxim Integrated Products: Hassan Rafat Mentor Graphics: * John Angulo * Arpad Muranyi Micron Technology: Randy Wolff Justin Butterfield QLogic Corp. James Zhou Andy Joy eASIC Marc Kowalski SiSoft: * Walter Katz * Todd Westerhoff * Mike LaBonte Synopsys Rita Horner Teraspeed Consulting Group: Scott McMorrow * Bob Ross (Note: Agilent has changed to Keysight) The meeting was led by Arpad Muranyi. ------------------------------------------------------------------------ Opens: - None -------------------------- Call for patent disclosure: - None ------------- Review of ARs: - Arpad write BIRD for double quote problem - Done - Ambrish investigate other ways to communicate tap coefficients - In progress - Arpad: Ambrish expects to give an update next week. - Arpad to review IBIS spec for min max issues. - In progress ------------- New Discussion: BIRD 128: - Arpad: This is on hold until the Friday open forum. - It is now scheduled for a vote in the next Open Forum. - Do we need to keep it on our agenda? - Radek: We need clarifications on some things: - Parameters to be included and excluded. - The possibility a DLL might read something uninitialized. - Walter: I proposed only a way to get "things" into AMI_GetWave. - It would be safest to have a NULL when nothing is in there. - I believe BIRD 147 calls for a string with the TX root and BCI data. - Radek: We might be opening the door for misinterpretation. - Walter: I would not object to having additional clarifications. - Arpad: Do we wait for the Open Forum to send this back to us? - Radek: We should be able to fix it now and communicate it to the Open Forum. - Arpad: We can make changes within a week and still have the vote. - Radek: Or the vote could be delayed if people register concerns. - Arpad: We do not have Ambrish today, he should be involved. - Radek: I might try to get it done by Friday. AR: Radek work with Walter and Ambrish on BIRD 128.1 clarifications. Double Quote BIRD draft: - Arpad showed QuoteCharBIRD_draft01.docx. - Arpad: This has a few concepts discussed here and in the Open Forum meeting. - There are restrictions on which files it applies to. - One section applies AMI files and another to DLL strings. - Arpad added "types" to the document. - Radek: It could say "all files covered by IBIS specification". - Arpad: We have a number of specifications. - John: We don't want to implicitly say DLLs need to follow ASCII rules. - Bob: ASCII does not distinguish between left and right double quote? - Arpad: Extended ASCII does, but not the original 7 bit ASCII. - The standard is ANSI X3.4-1986. - Walter: Touchstone can have binary files. - John: We can have "except as otherwise allowed". - Arpad: This is only about files in the IBIS standard. - Walter: I like John's suggestion. - Mike: "otherwise allowed by" what? - Arpad: We could explicitly exclude the types that don't apply. - Bob: That is too much information. - Mike: The IBIS specification can not make declaration about other specs. - Arpad: We have to say which file types this applies too. - Arpad: We might say it doesn't apply to Verilog, for example. - Mike: Does IBIS have general rules that apply to all of those file types? - Arpad: Not sure. - One problem is that some files do not have fixed extensions. - For example a package file does not have to be .pkg. - Bob: The checker searches for .pkg files automatically. - Arpad: I will leave this BIRD as is. AR: Arpad send double quote BIRD draft to Mike for posting AR: Mike post double quote BIRD draft MinMax BIRD draft: - Arpad showed MinMaxBIRD_draft01.docx. - Arpad: The keywords shown are scoped at the same level. - We might have to apply this rule at the component level. - Also we have Series keywords that might be independent. - Package and EBD might be independent too. - Mike: Could this just be an IBISCHK warning? - Arpad: The model maker should be aware. - Tools need to know how to treat the data. - Mike: We might have a [Component] keyword to say which corners are present. - Bob: [External Circuit] can make this messy. - We should have a rule that all columns are filled in. - We allow NAs but should not recommend it. - Arpad: We should be able to make models from lab measurements. - These have to be regarded as typical. - Bob: Highly recommending min and max would be good. - Walter: I would worry about getting bad data. - We could have a rule that if some are supplied then all must. - It should be limited to things that affect the IV characteristics of a buffer. - Bob: It should apply to all parameters, SubModel, etc.. - Arpad: Some keywords like Series ones are independent from the curves. - Bob: It would not include package keywords. - Arpad: Test Load poses additional problems. - Bob: Someone might use a min package with a min model, effectively mixing fast and slow. - Mike: If we do require min and max then they should not be identical to typical. - Radek: We have to allow for individual lines with NA in V-T and I-V tables. AR: Arpad send MinMax BIRD draft to Mike for posting AR: Mike post MinMax BIRD draft ------------- Next meeting: 02 September 2014 12:00pm PT ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives