[ibis-macro] Re: IBIS Interconnect Spice Dialect ...

  • From: "Walter Katz" <wkatz@xxxxxxxxxx>
  • To: <bradb@xxxxxxxxxxx>, "'IBIS Macro'" <ibis-macro@xxxxxxxxxxxxx>
  • Date: Fri, 5 Jun 2009 14:26:53 -0400

Brad,

 

I have submitted a proposal called Electronic Module Description (EMD),
which references Spice subckts. We put that aside to define the syntax or
SPICE dialect we would support in EMD. This was the Raison d'être for IBIS
Interconnect Spice. IBIS Interconnect Spice is really a sidebar for EMD. I
think we should be focusing on EMD and how we wrap SPICE subckts. As a
matter of fact, all of the interconnect subckts that we as EDA companies
create or use are de facto Hspice compliant,  and use the subset of Hspice
that are in the proposed IBIS Interconnect Spice Dialect. I am enclosing the
original presentation I made to IBIS ATM in March 2008. On slide 15 I
referred to ?IBIS Spice? Meta format (slide 15). Subsequently, IBIS ATM
agreed that a subset of HSPICE would be preferable to creating a new SPICE
dialect, and Synopsys graciously agreed to allow putting this subset of
their documentation into the public domain. 

 

Now that we have essentially agreed to the subset of HSPICE documentation we
would be using to define the IBIS Interconnect Spice Dialect, we need to
decide in what format we are going to publish this dialect.

 

I actually would want to poll the EDA companies sooner rather than later. I
think if all of the EDA companies will not support IBIS Interconnect Spice
Dialect today then I do not think we need to waste my time, or anyone else?s
time anymore on this and on EMD.

 

Walter 

 

-----Original Message-----
From: Brad Brim [mailto:bradb@xxxxxxxxxxx] 
Sent: Friday, June 05, 2009 1:25 PM
To: wkatz@xxxxxxxxxx; 'IBIS Macro'
Subject: RE: [ibis-macro] IBIS Interconnect Spice Dialect ...

 

hi Walter,

 

Thanks much for all your effort. This would not be moving so quickly without
all your mentoring.

 

You stated, "IBIS is planning to introduce new methods of modeling
connector, package and module interconnect."

Does the IBIS committee really have a documented short or intermediate term
goal of doing this? If not, then let's not make such statement here. In my
opinion, your paragraph is better without the first two sentences.

 

Stating in the proposal which EDA companies register support for it seems
like an early and potentially redundant vote/polling of the EDA companies.

 

best regards,

 -Brad

 


  _____  


From: ibis-macro-bounce@xxxxxxxxxxxxx
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Walter Katz
Sent: Friday, June 05, 2009 9:37 AM
To: IBIS Macro
Subject: [ibis-macro] IBIS Interconnect Spice Dialect ...

All,

 

I will be somewhere between Philadelphia and NYC during next Tuesday?s
meeting, so I may not be able to attend. I think the document I sent out
after this past Tuesday?s meeting is very close to fully defining the subset
of HSPICE that we intend to support. It requires careful review.

 

I think we are now at the stage to decide what formal final document we
would want to submit to the Open Forum to approve. I would like us to
consider making this document with out any significant changes as the
definition of the IBIS Interconnect Spice Dialect, and something like the
following to the IBIS Open Forum:

 

IBIS is planning to introduce new methods of modeling connector, package and
module interconnect. Each of these new modeling standards will require the
need to use SPICE subckts. Because of the large number of SPICE dialects,
IBIS ATM has determined the need for a ?least common denominator? SPICE
dialect called IBIS Interconnect Spice Dialect. Synopsys has given us
approval to use HSPICE documentation to define this dialect. The attached
document was generated bys cutting and pasting directly from the Synopsys
HSPICE manuals. A SPICE subckt is IBIS Interconnect Spice compliant if it
follows the syntactical rules described in this document. Below is a list of
EDA vendors that have agreed to support this SPICE dialect in subckts in
their EDA tools. The IBIS ATM committee hereby request that the enclosed
document be approved as the IBIS Interconnect Spice Dialect.

 

 

List of EDA Vendors registering support for this SPICE dialect.

Signal Integrity Software, Inc.

Synopsys 

?

 

I think that this approach can get this done very quickly, and we can move
on to using these SPICE subckts in package, connector and module models.

 

Walter

 

Walter Katz

Chief Scientist

Signal Integrity Software, Inc.

wkatz@xxxxxxxxxx

303.449-2308

 

Other related posts: