IBIS Macromodel Task Group

Meeting date: 16 July 2019

Members (asterisk for those attending):
ANSYS:                        Dan Dvorscak
                              Curtis Clark
Cadence Design Systems:       Ambrish Varma
                              Ken Willis
Intel:                        Michael Mirmak
Keysight Technologies:      * Fangyi Rao
                              Radek Biernacki
                              Ming Yan
                              Stephen Slater
                              Maziar Farahmand
Mentor, A Siemens Business: * Arpad Muranyi
Micron Technology:          * Randy Wolff
                              Justin Butterfield
SiSoft (Mathworks):         * Walter Katz
                              Mike LaBonte
SPISim:                     * Wei-hsing Huang
Teraspeed Labs:             * Bob Ross

The meeting was led by Arpad Muranyi.  Randy Wolff took the minutes.

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

- None

-------------
Review of ARs:
- Fangyi to write new BIRD draft for DC_Offset.
  - Done
- Arpad to submit [C Comp Model] BIRD to Open Forum.
  - Done.
    

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

- None.

-------------------------
Review of Meeting Minutes:

Arpad asked for any comments or corrections to the minutes of the July 9
meeting.  Randy moved to approve the minutes.  Bob seconded the motion. There
were no objections.

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

New DC_Offset BIRD draft from Fangyi:

Fangyi shared a PowerPoint "BIRD_enable_single_ended_signal_in_AMI.pptx". 
Motivation is the same as the BIRD197 DC_Offset BIRD.  DC_Offset usage is 
changed to In or InOut.  Defaults are 0 V. In the case of Usage In, the output 
value is assumed to be 0 V. 

Arpad asked about proposal item #1.  Is the default value of DC_Offset 0 
appropriate or should it default to the same as the input value?  This affects 
item #5. Fangyi doesn't want to add the input value to the output waveform.  
He doesn't want to mix input and output.

Fangyi showed two examples showing how the DC_Offset and NRZ_Threshold work 
together, along with multiple ways to use the parameters to get the same result.

Randy asked if these parameters would ever be used with differential SerDes 
buffers.  Arpad noted he was reminded of his question to the reflector a 
couple weeks ago about the use of the term threshold.  Fangyi said first 
assume Rx_receiver sensitivity is 0.  If you have a threshold, above the 
threshold is 1 and below the threshold is 0.  Arpad commented that we are 
looking only at a difference waveform.  Fangyi noted you should look at it 
with comparators.  Randy noted in SerDes we assume the threshold is 0.  Fangyi 
confirmed this, and Receiver_Sensitivity adds uncertainty to the switching 
level, ideally 0.

Arpad noted he sent comments by email to Fangyi.  Fangyi responded that he 
updated some descriptions in his presentation, but he needs to make the change 
in the BIRD draft.

Arpad stated it needs to be defined more clearly what happens when Get_Wave 
returns a value in DC_Offset.  Fangyi responded the only value that matters 
is the value returned by Init.  Arpad asked if GetWave would ever return the 
value.  Fangyi noted that GetWave could return the value, but only the value 
from Init is used to recover the waveform.

Wei-hsing asked about the GetWave flow.  Radek noted GetWave cannot handle 
InOut parameters.  This was discussed during the backchannel discussion.

Arpad asked if we should require the value to be returned only in Init and 
prohibit it elsewhere.  Fangyi thinks it doesn't matter if GetWave returns 
the DC_Offset value, since it is not used. 

Fangyi asked about Arpad's comment on saying "Rx equalizer output".  Fangyi 
noted an Rx may not have an equalizer.  Fangyi suggested "Rx algorithmic model 
output".  

Fangyi agreed with adding "may" to "EDA tools may obtain the complete waveform 
at the Rx output by" after Arpad described the EDA tool could process the 
waveform in different ways, so it is best not to dictate what the EDA tool 
should do.

Walter commented that he is ok with DC_Offset being In or InOut, and the rest 
of the BIRD is fine.

Bob asked if it would be helpful to put the examples in Fangyi's last two 
slides into the BIRD template.  Arpad suggested adding this in the background 
information section.

Bob asked if the intent was to supersede BIRD197.3.  Arpad wondered if it 
should be a revision to BIRD197.3 or a new BIRD.  Walter thought it would be 
easier to become a new draft of BIRD197.3, and he was ok with this becoming 
BIRD197.4.  

AR: Fangyi to create BIRD197.4 incorporating changes from his new BIRD and 
comments. 

There was a motion to adjourn and a second to the motion.  The meeting 
adjourned.


-------------
Next meeting: 23 July 2019 12:00pm PT
-------------

IBIS Interconnect SPICE Wish List:

1) Simulator directives