[ibis-macro] Minutes from the 08, 15, and 22 jan 2008 ibis-atm meetings
- From: "Mike LaBonte (milabont)" <milabont@xxxxxxxxx>
- To: <ibis-macro@xxxxxxxxxxxxx>
- Date: Mon, 28 Jan 2008 11:08:43 -0500
Minutes from the 08, 15, and 22 jan 2008 ibis-atm meetings are attached.
My apologies for getting behind on these.
Mike
IBIS Macromodel Task Group
Meeting date: 08 Jan 2008
Members (asterisk for those attending):
Ambrish Varma, Cadence Design Systems
* Arpad Muranyi, Mentor Graphics Corp.
Barry Katz, SiSoft
* Bob Ross, Teraspeed Consulting Group
* David Banas, Xilinx
* Donald Telian, consultant
Doug White, Cisco Systems
Essaid Bensoudane, ST Microelectronics
Ganesh Narayanaswamy, ST Micro
Hemant Shah, Cadence Design Systems
* Ian Dodd
Joe Abler, IBM
* John Angulo, Mentor Graphics
John Shields, Mentor Graphics
Ken Willis, Cadence Design Systems
Kumar, Cadence Design Systems
Lance Wang, Cadence Design Systems
Luis Boluna, Cisco
* Michael Mirmak, Intel Corp.
Mike LaBonte, Cisco Systems
Mike Steinberger, SiSoft
* Mustansir Fanaswalla, Xilinx
Patrick O'Halloran, Tiburon Design Automation
Paul Fernando, NCSU
Radek Biernacki, Agilent (EESof)
* Randy Wolff, Micron Technology
Ray Comeau, Cadence Design Systems
Richard Mellitz, Intel
Richard Ward, Texas Instruments
Sanjeev Gupta, Agilent
Shangli Wu, Cadence
Stephen Scearce, Cisco Systems
Steve Pytel, Ansoft
Syed Huq, Cisco Systems
Syed Sadeghi, ST Micro
* Todd Westerhoff, SiSoft
* Terry Jernberg, Cadence Design Systems
Vikas Gupta, Xilinx
Vuk Borich, Agilent
* Walter Katz, SiSoft
--------------------------
Call for patent disclosure:
- No one declared a patent.
-----
Opens:
- ATM API marketing (Don Telian)
- New toolkit/interoperability (Todd W.)
- There was a bug in the combined toolsets regarding a bad memory pointer
- Cadence addressed this but SiSoft's program made assumptions about
non-null returned information.
- Both companies are updating their kits.
- Three things needed:
- New SiSoft toolkit
- New Cadence toolkit
- A cookbook
- The latter will be published independently.
- No data from Cadence on when the update would be released.
-------------
Review of ARs:
- AMI BIRD authors meet to prove AMI BIRD can handle Mellitz cases
- Todd reported that he and SiSoft are debating about how to disclose
handling the linearity issues described in 2007 by Rich Mellitz.
- Mike S send non-proprietary information on studies he performed
- ???
- EDA vendors: meet to decide differential BIRD direction
- ???
- Michael M. sketch new IBIS interconnect plan
- ???
- Walter coordinate completion of true diff BIRD draft
- ???
- Michael M. draft Eye keywords for true diff BIRD
- TBD
- Randy draft derating keywords for true diff BIRD
- TBD
- Arpad: Write parameter passing syntax proposal (BIRD draft)
for *-AMS models in IBIS that is consistent with the
parameter passing syntax of the AMI models
- TBD
- TBD: Propose a parameter passing syntax for the SPICE
- [External ...] also?
- TBD
- Arpad: Review the documentation (annotation) in the macro libraries.
- Deferred until a demand arises or we have nothing else to do
-------------
New Discussion:
- Don: do we need an ATM-oriented support website to provide marketing push?
- Don represents a company that would like to add its logo and make an
announcement specific to the ATM API.
- Bob: can the company simply join the IBIS Open Forum?
- Should we have a press release?
- MM: GEIA supportive.
- Don mentioned that the public still views IBIS as "something less than
HSPICE."
- He also mentioned that this should be trumpeted a bit.
- Don's suggested next step: link to Matlab?
- Bob asked who the company was.
- Don deferred.
- Bob and Michael reminded the team that communications need to be open
and public.
- Ian cautioned being careful about communications regarding specific
companies.
- Todd: Changes online will take some time.
- We should get ATM out of macromodel_wip.
- Need easy-to-access link.
- Ian: where's EBD?
- Where did the website go?
- Bob and Arpad noted that the website has been restored and that EBD
definitions are part of the IBIS specification.
- Bob: We are moving away from [Diff Pin].
- Arpad: Don't mind [Diff Pin] etc. being left in the spec for pseudo-diff
- But not for true differential single-model-two-pins approach.
- To clean that up,we would need to introduce
- Bob: Are we specifying rules of evaluation or behavior?
- Currently, IBIS does not support true differential except under
[External Model].
- Todd: External model supports true differential?
- Walter/Bob: summarized state of current IBIS, measurements vs. behaviors
- Stimulus is also controlled by [Diff Pin]
- Walter: path 1 was to determine all the other measurement functions
through another set of keywords
- [Diff pin] and [Model Selector] cause issues
- Skew and other measurements should never have been placed at the pin level
- Bob: Agree; suggest Vdiff may be lined out
- Walter: Would like to limit one BIRD to measurement rules
- Having a parallel set of keywords to establish a cleaner structure
(Bob, Michael) would be separate.
- Walter: Keeping things clean would be a major problem
- Arpad: We should avoid having two avenues to achieve the same thing.
- Todd: behavioral descriptions and evaluation are two different things.
- Why not separate measurements and models?
- Bob, Michael and Arpad described how this idea has been proposed in the
past.
- Bob: Should we have old and new in the same file?
- Why this restriction?
- Doesn't the EDA tool have to have the same functionality?
Next meeting: 15 Jan 2007 12:00pm PT
-----------
IBIS Macromodel Task Group
Meeting date: 15 Jan 2008
Members (asterisk for those attending):
Ambrish Varma, Cadence Design Systems
* Arpad Muranyi, Mentor Graphics Corp.
Barry Katz, SiSoft
* Bob Ross, Teraspeed Consulting Group
* David Banas, Xilinx
Donald Telian, consultant
Doug White, Cisco Systems
Essaid Bensoudane, ST Microelectronics
Ganesh Narayanaswamy, ST Micro
Hemant Shah, Cadence Design Systems
* Ian Dodd
Joe Abler, IBM
* John Angulo, Mentor Graphics
John Shields, Mentor Graphics
Ken Willis, Cadence Design Systems
Kumar, Cadence Design Systems
Lance Wang, Cadence Design Systems
Luis Boluna, Cisco
* Michael Mirmak, Intel Corp.
* Mike LaBonte, Cisco Systems
* Mike Steinberger, SiSoft
Mustansir Fanaswalla, Xilinx
Patrick O'Halloran, Tiburon Design Automation
Paul Fernando, NCSU
* Radek Biernacki, Agilent (EESof)
* Randy Wolff, Micron Technology
Ray Comeau, Cadence Design Systems
Richard Mellitz, Intel
Richard Ward, Texas Instruments
Sanjeev Gupta, Agilent
Shangli Wu, Cadence
Stephen Scearce, Cisco Systems
Steve Pytel, Ansoft
Syed Huq, Cisco Systems
Syed Sadeghi, ST Micro
* Todd Westerhoff, SiSoft
* Terry Jernberg, Cadence Design Systems
Vikas Gupta, Xilinx
Vuk Borich, Agilent
* Walter Katz, SiSoft
--------------------------
Call for patent disclosure:
- No one declared a patent.
-----
Opens:
- DesignCOn presentations
- Todd is presenting on AMI interoperation
- Bob
-------------
Review of ARs:
- Michael M. : Summarize options/directions in a couple of for our
differential feature enhancements
- AMI BIRD authors meet to prove AMI BIRD can handle Mellitz cases
- Todd reported that he and SiSoft are debating about how to disclose
handling the linearity issues described in 2007 by Rich Mellitz.
- Mike S send non-proprietary information on studies he performed
- ???
- EDA vendors: meet to decide differential BIRD direction
- ???
- Michael M. sketch new IBIS interconnect plan
- ???
- Walter coordinate completion of true diff BIRD draft
- ???
- Michael M. draft Eye keywords for true diff BIRD
- TBD
- Randy draft derating keywords for true diff BIRD
- TBD
- Arpad: Write parameter passing syntax proposal (BIRD draft)
for *-AMS models in IBIS that is consistent with the
parameter passing syntax of the AMI models
- TBD
- TBD: Propose a parameter passing syntax for the SPICE
- [External ...] also?
- TBD
- Arpad: Review the documentation (annotation) in the macro libraries.
- Deferred until a demand arises or we have nothing else to do
-------------
New Discussion:
Mike S. presented spreadsheet NonlinearReconstruction.xls:
- Addressing slide 16 from Mellitz presentation.
- Measured data from slide with Kruler
- Shows that if tap weights other than -.21, .79 will yield close results.
- Device goes into saturation and fails to produce predicted amplitude.
- Only need to measure one pass.
- Need to measure for every EQ setting.
- Driver is non-linear but memory-less.
- There is a set of tap weights that will predict the non-linear behavior.
- Todd: engineering is about usable models, not necessarily perfection.
Michael M. presented "Advanced Differential Measurements for IBIS":
- Addresses an old AR: "Michael M. draft Eye keywords for true diff BIRD"
- Chart a direction for differential modeling
- Need to be specific about location Pin/Pad/Core
- Todd calls this "decision point"
- There are min and max V limits outside the eye.
- USB 2.0 spec chap 7 shows how it is defined
- Keywords:
- [Diff Receiver Measurements]
- [Eye Template] name
- Each has "bookend" keyword
- Can add more to [Diff Receiver Measurements] later
- UI column in point data is dimensionless, 0.0 to 1.0
- Walter: this is defining the lower & upper eyes together
- Could have separate columns
- Michael M: Might not allow for irregular UI axes
- Walter: Same eye patterns are used all the time
- Data would be repeated many times
- Should not have same eye templates in every file
- Would be nice to include these from a separate file
- Michael M: Should we get standards orgs to create these files?
- Todd: A simple IBIS .include mechanism would solve this
- Ian: Do we reference by file or by name?
- Bob: Might have trouble keep standard file name straight.
- Walter: JEDEC could supply these along with their specs
- Walter: There are multiple Core locations that Probe_location should address.
- External model may export nodes
- AMI model may export nodes
- Michael M: There is interest in latch-to-latch analysis
- Arpad: Walter had other parameters we might add
- Michael M: Maybe that is a separate effort
- If we start a new group we will probably have the same people
- Arpad: Would still prefer separate meetings
- Bob: Priority is to move forward quickly on what we have
- Walter: We should commit to a date for completing this BIRD, including:
- single-ended
- differential
- Tvac
- Tdelay
- Arpad: we need an AR
- Bob: Set "several months" as a target
- Walter: can have something after DesignCon
AR: Walter, Randy, Michael M. begin writing differential BIRD
Next meeting: 22 Jan 2007 12:00pm PT
-----------
IBIS Macromodel Task Group
Meeting date: 22 Jan 2008
Members (asterisk for those attending):
Ambrish Varma, Cadence Design Systems
* Arpad Muranyi, Mentor Graphics Corp.
Barry Katz, SiSoft
* Bob Ross, Teraspeed Consulting Group
David Banas, Xilinx
Donald Telian, consultant
Doug White, Cisco Systems
Essaid Bensoudane, ST Microelectronics
Ganesh Narayanaswamy, ST Micro
Hemant Shah, Cadence Design Systems
Ian Dodd
Joe Abler, IBM
* John Angulo, Mentor Graphics
John Shields, Mentor Graphics
Ken Willis, Cadence Design Systems
Kumar, Cadence Design Systems
Lance Wang, Cadence Design Systems
Luis Boluna, Cisco
* Michael Mirmak, Intel Corp.
* Mike LaBonte, Cisco Systems
Mike Steinberger, SiSoft
* Mustansir Fanaswalla, Xilinx
Patrick O'Halloran, Tiburon Design Automation
Paul Fernando, NCSU
Radek Biernacki, Agilent (EESof)
* Randy Wolff, Micron Technology
Ray Comeau, Cadence Design Systems
Richard Mellitz, Intel
Richard Ward, Texas Instruments
Sanjeev Gupta, Agilent
Shangli Wu, Cadence
Stephen Scearce, Cisco Systems
Steve Pytel, Ansoft
Syed Huq, Cisco Systems
Syed Sadeghi, ST Micro
* Terry Jernberg, Cadence Design Systems
* Todd Westerhoff, SiSoft
Vikas Gupta, Xilinx
Vuk Borich, Agilent
* Walter Katz, SiSoft
--------------------------
Call for patent disclosure:
- No one declared a patent.
-----
Opens:
-------------
Review of ARs:
AR: Walter, Randy, Michael M. begin writing differential BIRD
- Done
- Michael M. : Summarize options/directions in a couple of for our
differential feature enhancements
- Done
- AMI BIRD authors meet to prove AMI BIRD can handle Mellitz cases
- Todd reported that he and SiSoft are debating about how to disclose
handling the linearity issues described in 2007 by Rich Mellitz.
- Mike S send non-proprietary information on studies he performed
- Done
- EDA vendors: meet to decide differential BIRD direction
- Decided last meeting
- Michael M. sketch new IBIS interconnect plan
- In progress
- Walter coordinate completion of true diff BIRD draft
- In progress
- Michael M. draft Eye keywords for true diff BIRD
- In progress
- Randy draft derating keywords for true diff BIRD
- In progress
- Arpad: Write parameter passing syntax proposal (BIRD draft)
for *-AMS models in IBIS that is consistent with the
parameter passing syntax of the AMI models
- TBD
- TBD: Propose a parameter passing syntax for the SPICE
- [External ...] also?
- TBD
- Arpad: Review the documentation (annotation) in the macro libraries.
- Deferred until a demand arises or we have nothing else to do
-------------
New Discussion:
Walter presented on Receiver Measurements:
- Walter, Randy, and Michael M. met last week.
- Walter presented an outline
- Randy presented [T_Vac]
- Walter: slew rate units are not intuitive to work with
- Suggestion: Give voltage and use "/1n" as the units.
- Showing JESD79-3a spec
- Questions about interpolation method between table points
- JEDEC has Tvac discontinuity; jumps from 57ps to 75ps at 2.0V
- Arpad: this may be a JEDEC typo; 75ps should be 57ps
- Bob: These slew rates are within "groupings"
- Walter showed the "ac-ac or dc-ac" conundrum
- IBIS language for measuring slew in unclear
- Faster slew signals need to stay longer above Vin_ac
- The term "area" is used, but JEDEC does not measure that way
- Michael M.: These standards are commonly known by their bus names
- Communication of context for a measurement is important
- Michael M. presented A Selection of Common Mode Requirements in Industry Specs
- Discussion of USB 2.0 "Differential Input Voltage Range"
- Walter: hard to find commonality in measurement names used
- May need a cross reference across specs
- Michael M.: SATA measures at connector
- May have to define probe points there
- Maybe we could use "test load" for PCI-X
- PCI-X Vp-p has to be measured over 1e6 UIs
- DC is <30kHz
- Walter: If we find a well written standard we can use its terminology
- Next week we will discuss ICM links
- There will be no meeting Feb 5.
AR: Randy will investigate the JESD79-3a Tvac 57/75 "typo"
Next meeting: 29 Jan 2007 12:00pm PT
-----------
Other related posts:
- » [ibis-macro] Minutes from the 08, 15, and 22 jan 2008 ibis-atm meetings
