Minutes, IBIS Quality Committee

01 December 2015

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

ROLL CALL

eASIC:                              David Banas
Ericsson:                           Anders Ekholm
Intel:                              Michael Mirmak
                                    Eugene Lim
IO Methodology                    * Lance Wang
Signal Integrity Software         * Mike LaBonte
Teraspeed Labs:                   * Bob Ross

Everyone in attendance marked by *

NOTE: "AR" = Action Required.

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

Call for IBIS related patent disclosures:

- None

Call for opens:

- None

ARs:
- Mike update ibischk6 files online.
  - Done

- Mike send reflector notice about ibischk6
  - Delayed while we resolve bugs 168 and 169.

- Mike send testcase package model to Bob
  - Done, bug 168.

- Mike send W1503 testcase to Bob
  - Needs simplification first, has 986 lines.
  - Will send to Bob as-is for now.

- Mike add comments to CMPNT, DLY, MSPEC sections.
  - No new progress.

- Lance add comments to EBD section.
  - No report.

- Bob add comments to CIRCUIT section.
  - No new progress.

Emailed package model problem from AOTO:
- Bob: We need the IBIS file for this.
  - It may be that his component is missing power pins.

AR: Mike reply to AOTO about bug report

BUG168

- Mike: Some errors can be explained by printf issues.
  - Others are more serious.
- Bob: You tested a fix for this.
- Mike: It adds a 10 ppm tolerance.
- Bob: This is Vladimir's approach.
  - Simulators should tolerate small errors.
  - We have to check the file as-is.
  - EDA tools are free to clean up the files.
- Mike: Why do we have this rule?
- Bob: Perfect coupling is the case where the diagonal and off-diagonals are equal.
  - It is impossible for the off-diagonals to be larger in passive circuits.
- Mike: The other approach is a fixed tolerance.
  - It works with zero value diagonals, but maybe those should fail.
- Bob: A failing file is bad.
  - We might model this after what we do for non-monotonic curves.
  - Small infractions give a NOTE, ERROR for larger ones.
  - We will have to determine a small threshold.
  - There should be a minimum tolerance for the 186a case.
- Mike: The zero self-capacitance case seems impossible.
- Bob: We should print a message for that one.
  - Once we determine the sum of off-diagonals is greater than the diagonal:
    - Quick test: small diagonal values (close to zero) are a NOTE.
    - NOTE for violations under a 10 ppm tolerance.
    - ERROR for violations beyond a 10 ppm tolerance.
- Mike: This might be mentioned in ATM today.

AR: Bob update bug168.txt

AR: Bob update ibischk6 web page to note new bugs

AR: Mike send notice of IBISCHK6 with bug notes to email reflectors

IQ:

- Lance: The IQ2 level of the IBIS Quality Specification requires 4 V-T curves.
  - Models use different load values, a large R may be needed.
- Bob: Usually the V_fixture changes, not R_fixture.
  - And open drain/source will only need 2 waveforms.
- Lance: I can present more on this next time.

Meeting ended: 12:19 ET

Next meeting Dec 08, 2015
