IBIS Macromodel Task Group Meeting date: 12 Mar 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: - None -------------------------- Call for patent disclosure: - None ------------- Review of ARs: - Mike post updated BIRD 147 - Done - Arpad try to apply BIRD 158 shortcut language to his BIRDs 116-118 - In progress ------------- New Discussion: Interconnect Update: - Michael M: There was no meeting last week - The current agenda is still being worked on Arpad showed "Analog BIRD Review": - Michael M: The idea is to have relatively few BIRDs - Members are free to ask to schedule any of these BIRDs for a vote - Arpad: Is there a deadline to get into the next IBIS version? - Michael M: According to plan that would have been March 1 - We can still get the BIRDs approved will still help - Having unvoted BIRDs is the worst case - Only 1 BIRD is now approved - Bob: We need to make tough choices, including rejecting BIRDs - Mike L: The next release should be a permanent Open Forum agenda item - slide 2: - Arpad: Most of these are my BIRDs - 116.1, 117.4, 118.3, 129, 145.2, 153 - Arpad went over the status of each BIRD in detail - Arpad: They need to be converted to the new format - Then I would like to vote to vote on them - Walter: [External Circuit] was targeted for package interconnect, not on-die - BIRD 145 should not go through - It is fundamentally flawed - Arpad: BIRD 145 is not absolutely necessary - Do not agree the idea is fundamentally flawed - Are you saying BIRD 116 should be removed too? - Walter: I am only saying we should not work on [External Circuit] - Not even to add IBIS-ISS to it - Arpad: It's not a complicated addition - Bob: Technically we already support on-die interconnect with [External Circuit] - BIRD 153 should be moved to a separate file - Arpad: We should all look these BIRDs over and be ready to vote - Ambrish: Should we discuss BIRDs 116-118 before converting to new format? - Michael M: Keywords are added to the spec, but there is no requirement to support them - No one has used transit time, for example - Radek: This does not end the analog model issue - In [External Model] everything has to be in IBIS format - It is a separate issue whether IBIS-ISS will be usable - Arpad: These BIRDs may not address every problem - But they address enough - Meanwhile we can work on the long term solution - Walter: [External Circuit] does not come close to solving the analog model problem we have today - It would be a waste of time - Arpad: It is just adding a language, not a waste of time - Bob: The BIRDs should reference IBIS 5.1, not 5.0 - Once this is done we need a parser for it - That is the best way to adopt these BIRDs - It will take funding to get this done - Arpad: It is a short term solution - Bob: It might take 2 years - Michael M: Next OF meetings are this Friday and Apr 15 - Keep tat in mind if we want to set a new deadline date Fangyi showed "Redriver Background" (from recent DesignCon): - http://www.eda.org/ibis/summits/jan13/bari.pdf - slide 7: - Fangyi: In a redriver there is no retiming - It can be non-linear - It can't go in the current flow - Arpad: For very long channels can many of these be cascaded? - Fangyi: Yes, this BIRD supports that - slide 8: - Fangyi: It has to be modeled to behave like an analog circuit - It is driven by an analog waveform, not an event - The redriver TX is different from a regular TX - Michael M: Is this limited to bit by bit time domain simulation? - Fangyi: This is focused on time domain - Statistical is possible, but there are practical issues - For example, how to model device internal noise - Todd: Does the BIRD say it is for time domain? - Fangyi: No - slide 7: - Michael M: Isn't jitter amplified? - Fangyi: We assume a linear, noiseless, and passive channel - Device noise must be translated to jitter somehow - slide 10: - Fangyi: The TX and RX halves are in the same IBIS file - They have separate AMI and DLL files - Bob: Should the Model_type indicate they are redrivers? - Michael M: This gives no indication the RX feeds to the TX - Fangyi: We are open to suggestions about that - Todd: Walter proposed a means for that some time ago - Fangyi: There is no conflict with Walter's proposal - Ambrish: We should support non-AMI redrivers too - Fangyi: This BIRD assumes AMI - Walter: My [Repeater Pins] keyword could be [Buffer Pins] - It would be a retimer or redriver depending on whether there is an AMI model, and whether it returns clock_times - Ambrish: We need to go into the flow of how that works - Radek: As written it is only valid for AMI simulations - Bob: All of the analog effects are handled in the DLL - The TX part would drive the downstream channel - Fangyi: But it is pure analog, not event driven - Michael M: The only parallel we have is [Series MOSFET] Arpad: What needs to be done to these BIRDs 131 and 158? - Ambrish: There should be no problem merging them - Fangyi: I would prefer to keep them separate - Bob: Should we vote one down? - Ambrish: We can't vote down 131 - Walter: I will create a new BIRD to propose the pin keywords AR: Walter write new BIRD for repeater/retimer pin keywords ------------- Next meeting: 19 Mar 2013 12:00pm PT Next agenda: 1) Task list item discussions ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives