Minutes from 01 nov 2005 meeting attached. ARs due: AR: Mike fix EQ amp rcvr template AR: Ian will draft a BIRD for analog-only AMS support in IBIS AR: Mike add extrapolation warning to module documentation I will be out next Tuesday, unable to attend. If everyone else decides to carry on without me, we just need someone to take minutes. Mike
meeting date: 01 nov 2005 attending: Arpad Muranyi, Ian Dodd, Todd Westerhoff, Mike LaBonte Bob Ross, Ken WIllis, Shang Li ------------- Review of ARs: AR: Mike will annotate library modules - XML files, PERL scripts, almost done AR: Ken write simple receiver template - done AR: Ian will draft a BIRD for analog-only AMS support in IBIS - In progress ------------- Simple receiver - Based on EQ amp rcvr template AR: Mike fix EQ amp rcvr template Draft BIRD - Ian working with Arpad - Will support Verilog-AMS, but allow for A/D and D/A - specify required lo/hi V, and dt for analog inputs - lo/hi V for analog output - Discussion with Mentor analog sim people - No Verilog support in SI tools - Positive feedback so far - Analog portion of Verilog is fuzzy - 1999 standard, System-Verilog, etc. - Ken unable to contact Cadence people yet - Positive feedback so far - Current macro library would be illegal under current IBIS - BIRD probably ready in 2 weeks AR: Ian will draft a BIRD for analog-only AMS support in IBIS PWL extrapolation policy - SE = slope extrapolation - HE = horizontal extrapolation - extrapolation option: - On gives SE - Off gives error messages - V/T tends to be HE, everything else SE - Can always flatten the ends in case SE is used - flexibilty vs. compatibility - Use HE for dependent-V, SE for all else? - no, some dependent sources not ready for that - IBIS parser assumes SE of I/V curves - PWL VA code could check last 2 points at initialization - For tool compatibility - Not converging on this ... AR: Mike add extrapolation warning to module documentation Param arrays - Cadence simulation is problematic - HSPICE has same problem - Re-write so tools can consume directly? - Can cut & paste inot model instead of using external - PERL script could do it Triggerable source - Earlier decision was to do this later - Arpad running into it now - Tlsim is the only implementation ------------- Next meeting: Tuesday 08 nov 2005 (Mike can't make it).