[ibis-macro] Re: What I think BIRD145 is try to accomplish is ...

  • From: Taranjit Kukal <kukal@xxxxxxxxxxx>
  • To: "wkatz@xxxxxxxxxx" <wkatz@xxxxxxxxxx>, "'IBIS-ATM'" <ibis-macro@xxxxxxxxxxxxx>
  • Date: Thu, 16 Feb 2012 10:11:38 +0530

Hi Walter,
Attached presentation summarizes BIRD145 proposal -

rgds
..kukal


________________________________
From: ibis-macro-bounce@xxxxxxxxxxxxx [mailto:ibis-macro-bounce@xxxxxxxxxxxxx] 
On Behalf Of Feras Al-Hawari
Sent: Thursday, February 16, 2012 1:55 AM
To: wkatz@xxxxxxxxxx; 'IBIS-ATM'
Subject: [ibis-macro] Re: What I think BIRD145 is try to accomplish is ...

Walter,

Complete examples and details are shown/discussed in BIRD 145, please go over 
it and let me know if you have further questions.

In summary, if you want to develop Example 2 in BIRD 145 based on the current 
IBIS spec you need to:
- Replace the [Model] block with an [External Circuit] block in Figure 2 in 
BIRD 145
- Develop IBIS/ISS or SPICE-wrappers (for the typ/min/max corners) in which the 
corresponding [Model] is instantiated
- Add an [External Circuit] section that corresponds to the [Model] block and 
also points to the SPICE-wrapper subcircuits



On the other hand, BIRD 145 allows us to directly call an IBIS I/O [Model] and 
then cascade it to other [External Circuit] blocks using the new [Model Call] 
keyword as shown in Examples 1 and 2 in BIRD 145.



Feras






________________________________
From: Walter Katz [wkatz@xxxxxxxxxx]
Sent: Wednesday, February 15, 2012 6:56 AM
To: Feras Al-Hawari; 'IBIS-ATM'
Subject: RE: What I think BIRD145 is try to accomplish is ...

Feras,

I do not understand your response. Can you put together a complete example of 
what you are proposing?

Walter

From: Feras Al-Hawari [mailto:feras@xxxxxxxxxxx]
Sent: Wednesday, February 15, 2012 4:28 AM
To: wkatz@xxxxxxxxxx; IBIS-ATM
Subject: RE: What I think BIRD145 is try to accomplish is ...

Hello Walter,

The main goal of BIRD 145 is to directly cascade and call an IBIS I/O model 
(standard VI/VT based model or an [External Model]) with an [External Circuit] 
that could be a represenation of ODT, RDL, or package. Based on that, for 
example, the [External Circuit] could reference an IBIS/ISS subckt that 
contains your suggested ladder RC circuit.

So BIRD 145 would allow you to directly connect your ladder RC circuit 
(encapsulated inside the [External Circuit]) to its corresponding buffer model 
using the new [Model Call] subparameter. To accomplish the previous goal based 
on the current IBIS spec you need to: 1) encapsulate the b-element inside a 
subckt, 2) reference that b-element subckt from an [External Circuit], 3) 
connect [External Circuit] that references the I/O model to the [External 
Circuit] that references the ladder RC circuit. It is obvious that these steps 
are cumbersome especially when the developer could just directly call the 
desired IBIS I/O model without the need to develop a SPICE-like wrapper as well 
as an [External Circuit] section.

Based on the above, the goal of BIRD 145 is different from your proposal  below.

Thanks,

Feras

________________________________
From: Walter Katz [wkatz@xxxxxxxxxx]
Sent: Tuesday, February 14, 2012 6:24 PM
To: IBIS-ATM; Feras Al-Hawari
Subject: What I think BIRD145 is try to accomplish is ...
Feras,

I think the idea behind BIRD145 is similar to what Michael Mirmak proposed 
several years ago, and what Randy Wolf discussed at the DesignCon IBIS Summit. 
Both Randy and Michael wanted the ability to add a ladder RC circuit at the 
output pad of the buffer. This is a real requirement that SiSoft has 
implemented using an IBIS keyword Final_Stage, FinalStageDiff:

Final_Stage <typ|min|max> <ISS file name> <ISS subckt name>
and
Final_Stage_Diff <typ|min|max> <ISS file name> <ISS subckt name>

These are either a 2 port subckt (Final_Stage) or a 4 port subckt 
(Final_Stage_Diff), that is inserted between the pad(s) of the B element and 
the pad end(s) of the package).

Would this satisfy your IC Vendor modeling requirements? What other ports would 
need to be included in this subckt?

Walter

Walter Katz
wkatz@xxxxxxxxxx<mailto:wkatz@xxxxxxxxxx>
Phone 303.449-2308
Mobile 720.333-1107

Other related posts: