Minutes, IBIS Quality Committee

30 November 2010

11:00-12:00 EST (08:00-09:00 PST)

ROLL CALL

Cisco Systems:                  Mike LaBonte*
Ericsson:                       Anders Ekholm*
Green Streak Programs:          Lynne Green
Huawei Technologies:            Guan Tao
IBM:                            Bruce Archambeault
IOMethodology:                  Lance Wang*
Mentor Graphics:                John Angulo
Micron Technology:              Moshiul Haque, Randy Wolff
Nokia Siemens Networks:         Eckhard Lenski*
Signal Consulting Group:        Tim Coyle
Teraspeed Consulting Group:     Bob Ross*
Texas Instruments:              Pavani Jella

Everyone in attendance marked by *

NOTE: "AR" = Action Required.

-----------------------MINUTES ---------------------------
Mike LaBonte conducted the meeting.

Call for opens and IBIS related patent disclosures:

- No one declared a patent.

AR Review:

- Moshiul update IQ checklist to add |IQ column
  - Done
  - Mike showed the checklist and fixed a font issue live
  - This will be posted to the website

- Mike post updated External Test Data and Test Load BIRD
  - This was sent to the list by email

AR: Mike post updated IQ checklist
AR: Mike post External Test Data and Test Load BIRD draft

New items:

New External Test Data and Test Load BIRD draft:
- Anders: We should try some real test circuits first
  - Also, how do we specify drivers for testing inputs?
  - We are probably not capturing this accurately
- Mike: Testing inputs is important, especially with C_comp issues
  - Maybe we need [Test Source]
  - IBISCHK would have to not call [Test Source] buffers unreferenced
- Bob: IBISCHK 4 has that fixed
- Mike showed the original Golden Waveform BIRDs
  - BIRD 69.1 was rejected
    - This one combined load and data into one keyword
  - BIRD 70.5 was accepted
- Bob: There is no absolute time in IBIS
- Mike: Maybe only the test source voltage waveform needs to be captured
- Anders: That may not be the case
  - The reflection response may matter
- Bob: A single pulse would be no problem
- Anders: We should test multiple pulses
- Bob: There are three solutions:
  - PULSE language
  - PWL language
  - A SPICE language like IBIS-ISS that supports sources
- Mike: PWL would be good to eliminate the IBIS driver as an error source
  - It is best that only the DUT is questionable
  - Using ideal drivers would be better for testing inputs
- Bob: It would not model the source impedance for reflection effects
- Mike: If we capture the voltage it should already have that reflection
  effect in it
- Anders: It will depend on the simulator
  - The simulators may differ in how they handle the T-lines
- Mike: Then IBIS-ISS might have a problem in all of it's uses

AR: Mike test source voltage capture idea

Anders: We need to think about some cases we would like to test
- Bob: We want tests that actually show something
  - JEITA uses 12 test loads, for example
- Anders: We may not have that many, but we need to see what kinds we will have
- Bob: We also need to consider typ/min/max

Bob: We need EDA tools that will make running these tests easier
  - But it's not their core business
- Mike: Maybe a 3rd party would implement it
- Bob: It has to be simple and straightforward

Next meetings:
- Next meetings Dec 7 and Dec 14

Dec 7 agenda:
- New External Test Data and Test Load BIRD draft

Meeting ended at 12:09 Eastern Time.