IBIS Macromodel Task Group

Meeting date: 11 March 2014

Members (asterisk for those attending):
Agilent:                    * Fangyi Rao
                              Radek Biernacki
Altera:                       David Banas
ANSYS:                      * Dan Dvorscak
                            * Curtis Clark
Cadence Design Systems:     * Ambrish Varma
                            * Brad Brim
                            * Kumar Keshavan
                              Ken Willis
			      Scott Huss
Ericsson:                     Anders Ekholm
Intel:                      * Michael Mirmak
Maxim Integrated Products:    Hassan Rafat
Mentor Graphics:            * John Angulo
                            * Arpad Muranyi
			      Andrey Matvienko
Micron Technology:          * Randy Wolff
                              Justin Butterfield
QLogic Corp.                  James Zhou
                              Andy Joy
SiSoft:                       Walter Katz
                              Todd Westerhoff
                            * Mike L LaBonte
Teraspeed Consulting Group:   Scott McMorrow
                            * Bob Ross

The meeting was led by Arpad Muranyi

------------------------------------------------------------------------
Opens:

- Arpad: Walter is on vacation and will not join us today.

--------------------------
Call for patent disclosure:

- None

-------------
Review of ARs:

- Mike L post BIRD 147 draft 7
  - Done

- Bob and Ambrish - BIRD 147 editorial work.
  - Done

- Arpad produce examples for BIRD 163, 164, 165
  - In progress.

-------------
New Discussion:

BIRD 147 draft 7:
- Ambrish showed the BIRD.
- Ambrish: Bits have been simplified, only binary is accepted.
- Arpad: It is OK to remove octal, but hex is pretty useful.
- Kumar: It would be better to avoid complexity.
- Mike L: With hex policies are need for when the number of bits is not a multiple of 4.

- Ambrish: The LFSR section has been extended with more information about how to generate.
- Arpad: It should define which end of the bits is most significant.

- Mike L: Can anything but "BCI" be under Protocol_Specific?
- Ambrish: That would not be allowed.
- Kumar: We might just replace Protocol_Specific with BCI.
- Ambrish: Protocol_Specific gets stripped out.
- Mike L: We might replace BCI with protocol names and allow multiple protocols.
- Kumar: It is set up for one protocol at a time.
- Mike L: Could the contents under BCI be raised up a level?
  - They are like the items right under Model_Specific.
- Ambrish: There is a Training_Done parameter that is special.
- Mike: There should be a rule that nothing but BCI can appear, so it will be an IBISCHK error.
- Ambrish: Agree

- Arpad: Will this need more discussion?
  - Maybe we should discuss BIRD 128 before voting on this.

Arpad: The interconnect task group is resuming.
- Michael M: It begins not tomorrow but next Wednesday.
- Arpad: The time is the same as before.

Arpad showed the Tabled BIRDs list and reviewed the items:
- Michael M motioned to untable item 9, bug 151.
- Michael M: The bug is specifically about single ended support.
  - The IBIS 6.0 says nothing about directionality for algorithmic models.
  - An AMI model assigned to single ended buffers would create a directionality problem.
- Arpad: I Don't think AMI can be assigned to single ended.
- Michael M: There is no explicit restriction.
  - On page 167 there is a phrase that seems to allow single ended.
  - There is a single waveform, for one thing.
- Arpad: Should we have a BIRD?
- Michael M: This has been discussed a bit.
  - We may need a .ami file parameter for this.
  - If someone connected RX to RX how would the tool know?
- Arpad: Regular buffers have enable controls, how would we know not to use it?
- Ambrish: We usually define this in the [Model].
- Michael M: It think Model_Type I/O works for AMI right now.
  - That could cause some problems if the tool used it.
- Ambrish: All AMI models take a wave in and put a wave out.
- Arpad: It is up to users to use these in the right place.
- Michael M: I might want TX and RX behavior in one DLL.
  - We need to say what direction should be applied.
- Arpad: We could have an enable signal passed into the DLL.
- Michael M: Walter had suggested a Model_Selector to handle this.
  - An I/O buffer could have a selector that makes it only I or O.
- Arpad: There could be multiple [Algorithmic Model] under the [Model].
- Michael M: There would be only one analog model.
  - Model_Selector could work now.
  - A pin with an enable control might want different filter behavior for each mode.
- Ambrish: Could we require TX and RX in the names?
- Michael M: An I/O buffer would be both.
- Arpad: We should have an enable signal for that.
- Bob: For I/O the user currently has to choose a direction to use.
  - This would be the same.
- Michael M: An [Algorithmic Model Selector] would make the most sense.
  - The changes would be small.
- Bob: There should be a Reserved_Parameter for directionality.
- Arpad: That would be best.
- Michael M: We would need that and [Algorithmic Model Selector].
- Arpad: It would be best to have it so no user intervention might be required.

-------------
Next meeting: 25 March 2014 12:00pm PT

-------------
IBIS Interconnect SPICE Wish List:

1) Simulator directives