[ibis-macro] EBD Modeling was origionaly meant to describe packages ...

  • From: "Walter Katz" <wkatz@xxxxxxxxxx>
  • To: <Arpad_Muranyi@xxxxxxxxxx>, "'IBIS-ATM'" <ibis-macro@xxxxxxxxxxxxx>
  • Date: Wed, 15 Aug 2012 13:15:56 -0400 (EDT)

Arpad,

 

Going back to BIRD 36.2:

 

STATEMENT OF THE ISSUE:  There is a need to describe SIMM modules and
related

type components that consist of one or more ICs mounted on a PCB board
that

connects them to a system thru a set of pins.  The following BIRD proposes
a

new type of file called .ebd (Electrical Board Description) that

addresses this need.  This proposal does not encompass an electrical 

description of connectors and other interconnect devices.

 

A SIMM is a type of memory package
(http://computer-building-repairs.knoji.com/types-of-memory-packages

 

So how can you say that EBD was not meant to describe packages?

 

I am enclosing three presentation that I gave 4 years ago about Electronic
Module Design (EMD). These presentations were put on hold because we
needed IBIS-ISS as the language to describe interconnect in a Module.

 

EMD was designed to handle general boards as well as what we currently
call EBD, Package, MCM. The committee made a clear decision at that time
that "Boards" and "Connectors Between Boards" was the real of the EDA
companies, and should not be made an IBIS standard. 

 

If you peruse these documents you can see that the latest proposals I have
made are essentially the same concepts as what I had proposed 4 years ago,
but with different syntax and some additional capability to reflect what I
have learned in the last 4 years. 

 

Walter

 

From: ibis-macro-bounce@xxxxxxxxxxxxx
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Muranyi, Arpad
Sent: Wednesday, August 15, 2012 12:29 PM
To: 'IBIS-ATM'
Subject: [ibis-macro] Re: Straw Poll vote on IBIS-ISS Package Modeling

 

Walter,

 

I knew you will comment on the word "substrate" .  J

 

The problem is that this word is not defined in the IBIS

specification.  When I first learned this word, it meant

the silicon substrate of the die on which the various

layers were deposited to make the integrated circuit.

 

Then came the BGA-s (and MCM-s) and those tiny green

printed circuit boards which carried the black plastic

molding of the package that encapsulated and protected

the die.  When I first saw these, I considered the green

tiny plates printed circuit boards, but they started to

call them substrates.  First time I heard that, I was

really surprised because I thought substrate was part

of the die.

 

Now you say that substrate is the package.

 

This is a good illustration how a word can mean different

things at different times in history.

 

Here is a definition from an online dictionary:

 

http://www.thefreedictionary.com/substrate

 

I really like #3 in this definition, which basically says

that a substrate is a substratum.  That tells me a lot.  J

Well to be fair, it also says that it is an underlying layer.

By this definition, we can start calling a printed circuit

board a substrate also because it is an underlying layer

below all the components which are soldered onto them.

 

Either way, this is why it is so hard to write a specification.

Even in a decade or so time, the meaning of words can change

and then we get into these discussions.  But reading the word

in context, I still maintain that when we wrote EBD we were NOT

thinking about package modeling.  Another point that supports

this reasoning is that the EBD syntax is almost identical but

inferior to the .pkg syntax in IBIS.  In EBD we didn't include

the RLC matrix syntax, only the T-line syntax because we were

thinking about transmission lines on printed circuit boards.

RLC matrices and bond wires, etc. seemed to be more applicable

for package modeling in the days when EBD was conceived and

written, and the fact we left these out from EBD indicates to

me that we didn't intend to use it for package modeling...

 

Anyway, I don't want to dwell on history.  I would just like to

have a package modeling syntax that can do stacked die also.  I

don't think we should chop things up so that we would need three

different files (.ibs, .pkg, and .ebd) for stacked die package

models.  I would prefer to be able to do that in the package

modeling syntax and .pkg file.  I think it would be cleaner and

less confusing that way.

 

Thanks,

 

Arpad

================================================================

 

 

 

 

From: Walter Katz [mailto:wkatz@xxxxxxxxxx] 
Sent: Wednesday, August 15, 2012 10:49 AM
To: Muranyi, Arpad; 'IBIS-ATM'
Subject: RE: [ibis-macro] Re: Straw Poll vote on IBIS-ISS Package Modeling

 

Arpad,

 

| A "board level component" is the generic term to be used to describe a 

| printed circuit board (PCB) or substrate

 

A substrate is a package.

 

The following statement may have been the motivation for EBD, but this
text does not appear in IBIS 5.0

-    it was designed to describe the printed circuit BOARDs

of memory modules in the days when memory modules used to

be modeled as a lumped capacitance and we felt that more

details were necessary to find the timing to the memory

chips on these memory modules

 

The common usage of EBD files includes memory Dimms and Multi-Chip Modules
(MCM). The example is an MCM package which contains two memory packages
that are stacked.

 

I am enclosing three files:

1.       Legacy.ebd is extracted from a real memory EBD file representing
a package currently being delivered by an IC Vendor. I redacted the name
of the Vendor. This is a subset of the pins in the original EBD file
containing 3 DQ (single ended data), one DK (differential clock), and one
A (single ended address). This package contains two memory packages which
are stacked. The memory packages themselves are Small Outline Packages
(SOL) which will have their own package models.

2.       IBIS-ISS.ebd is this same package rewritten using my proposed
syntax. I include Gd and Rs on all of the wline models (the IC Vendor
added these as comments in the Legacy.ebd file).

3.       IBIS-ISS_Tstonefile.ebd is this same package rewritten using my
proposed syntax using Touchstone models. The DK is done as a differential
model. I added a Power subckt and a coupled model for one of the DQ
models.

 

Features that are left as an exercise for the user:

Full package models. 

Interface package models. 

Bank package models. 

Coupling between signal and power.

 

I do not think that IBIS-ISS package need to support the "Model_name"
feature that we need to support in .ibs files.

 

Walter

 

From: ibis-macro-bounce@xxxxxxxxxxxxx
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Muranyi, Arpad
Sent: Wednesday, August 15, 2012 10:23 AM
To: 'IBIS-ATM'
Subject: [ibis-macro] Re: Straw Poll vote on IBIS-ISS Package Modeling

 

Walter,

 

I would disagree with "This is exactly what EBD was designed for".

-    its name says Electrical BOARD Description

-    it was designed to describe the printed circuit BOARDs

of memory modules in the days when memory modules used to

be modeled as a lumped capacitance and we felt that more

details were necessary to find the timing to the memory

chips on these memory modules

-    the introduction to Section 8 on pg. 165 of the v5.0 spec

tells everyone very clearly what its purpose is:

 

 

|=========================================================================
==== 

|=========================================================================
==== 

|                       Section 8 

| 

| E L E C T R I C A L   B O A R D   D E S C R I P T I O N 

| 

|=========================================================================
==== 

|=========================================================================
==== 

| 

| A "board level component" is the generic term to be used to describe a 

| printed circuit board (PCB) or substrate which can contain components or


| even other boards, and which can connect to another board through a set
of 

| user visible pins. The electrical connectivity of such a board level 

| component is referred to as an "Electrical Board Description". For
example, 

| a SIMM module is a board level component that is used to attach several
DRAM 

| components on the PCB to another board through edge connector pins. An 

| electrical board description file (a .ebd file) is defined to describe
the 

| connections of a board level component between the board pins and its 

| components on the board. 

| 

| A fundamental assumption regarding the electrical board description is
that 

| the inductance and capacitance parameters listed in the file are derived


| with respect to well-defined reference plane(s) within the board. Also, 

| this current description does not allow one to describe electrical 

| (inductive or capacitive) coupling between paths. It is recommended that
if 

| coupling is an issue, then an electrical description be extracted from
the 

| physical parameters of the board. 

 

 

Whether EBD is "being used for, and are being delivered for" stacked die
or

multi-chip modules is a different question.  You can use fertilizers

to grow plants, but you can use them to make bombs also.

 

Thanks,

 

Arpad

=====================================================================

 

From: Walter Katz [mailto:wkatz@xxxxxxxxxx] 
Sent: Wednesday, August 15, 2012 7:14 AM
To: Muranyi, Arpad; 'IBIS-ATM'
Subject: RE: [ibis-macro] Re: Straw Poll vote on IBIS-ISS Package Modeling

 

Arpad,

 

The following statements are totally incorrect:

 

Using EBD as a package to house multiple components (each of

which describes a packaged die) looks a little strange to me,

even if the component's package model is zeroed out.

 

For this reason I think that using EBD for multi-chip package

modeling is NOT the right thing to do.  It simply doesn't fit

the IBIS hierarchy.

 

This is exactly what EBD was designed for, and are being used for, and are
being delivered for.

 

Walter

 

 

From: ibis-macro-bounce@xxxxxxxxxxxxx
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Muranyi, Arpad
Sent: Wednesday, August 15, 2012 1:44 AM
To: IBIS-ATM
Subject: [ibis-macro] Re: Straw Poll vote on IBIS-ISS Package Modeling

 

Hello everyone,

 

I feel that it would be a little too early to take a vote with

these questions, because it seems that there are a few things

we need to clear up first.  Here is one issue:

 

Randy commented today that he would like to have multi-chip

capabilities within a package.  This is not listed in the

questions.  I do remember Walter's answer to that comment

being that we can do it with EBD, but I am not so sure about

that.  It seems that we have a hierarchical problem there.

 

Looking at "Section 3a" in the v5.0 specification on pg. 13

I see [Package] and [Package Model] under [Component].  This

indicates that [Component] doesn't refer to a die, but it

refers to a packaged die.  In order to have multiple, possibly

different dice in the same package, we would need to have 

another keyword which is lower than [Package] or [Package Model]

in the hierarchy for the die.  Currently we don't have this.

The IBIS specification seems to be architected with the one

die per package assumption.

 

Looking at the architecture of the EBD specification, I see

the [Reference Designator Map] keyword which is responsible

to associate any reference designators (U1, U2, etc.) listed

in any "Node" entries of the .ebd file with an .ibs file and

component name.  The syntax in EBD is RefDes.PinName where 

PinName is the pin name of a component.  This clearly assumes

that a component that is referenced by the reference designator

of EBD is a packaged die (since it is using a pin name, not a

pad name).

 

Using EBD as a package to house multiple components (each of

which describes a packaged die) looks a little strange to me,

even if the component's package model is zeroed out.

 

For this reason I think that using EBD for multi-chip package

modeling is NOT the right thing to do.  It simply doesn't fit

the IBIS hierarchy.

 

Comments, questions welcome.

 

Thanks,

 

Arpad

================================================================

 

From: ibis-macro-bounce@xxxxxxxxxxxxx
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Walter Katz
Sent: Tuesday, August 14, 2012 4:19 PM
To: IBIS-ATM
Subject: [ibis-macro] Straw Poll vote on IBIS-ISS Package Modeling

 

All,

 

I have put 28 questions in a spreadsheet. I think it will take too much
time to do the voting in the meeting, so I request that you put entries
1:4 in column 1 of each question. The meanings of 1:4 are.

 


4

Advocate   I need this!


3

Support      I support this


2

Abstain       I have no opinion


1

Object         I object to this functionality

 

Return them to me or all IBIS-ATM and I will collate the results and
publish them by Tuesday AM. I am including the .xlsx file in case you have
difficulty editing the spreadsheet in the body of this e-mail. Also
including this and last weeks presentation.

 

Walter

 


4

Advocate   I need this!


3

Support      I support this


2

Abstain       I have no opinion


1

Object         I object to this functionality


 

 


 

Should we add keywords Pullup_Signal_name, Pulldown_Signal_name,
Power_clamp_Signal_name, and Ground_clamp_Signal_name to the [Model]
section?


 

Should we add a section to the .ibs file to define the voltage values of
supply signal names?


 

Should we add a list of supply die pad names?


 

Should we add an x-y coordinate for each pin and die pad?


 

 


 

Should we support two signal pins connected to the same die pad (Forked
Signal)?


 

Should we be able to associate a package model with a [Model]?


 

Should we be able to associate a package model with a Pin_name?


 

Should we be able to associate a coupled package model with a [Model]?


 

Should we be able to associate a coupled package model with a list of
Pin_names?


 

Should we support package models with coupling between signals and power?


 

Should we support a coupled package model that hooks up to two or more
[Model] names?


 

Should we support package models with more than 3 corners?


 

Should we support package Touchstone files directly?


 

Should we support sparse usage of large package Touchstone files?


 

Should we support package "Quadrants" (e.g. Banks, Interfaces)?


 

Should we support full package models?


 

 


 

Should we support on-die models?


 

Should we be able to associate an on-die model with a [Model]?


 

Should we be able to associate an on-die model with a Pin_name?


 

Should we be able to associate a coupled on-die model with a [Model]?


 

Should we be able to associate a coupled on-die model with a list of
Pin_names?


 

Should we support on-die models with coupling between signals and power?


 

Should we support a coupled on-die model that hooks up to two or more
[Model] names?


 

Should we support on-die models with more than 3 corners?


 

Should we support on-die Touchstone files directly?


 

Should we support sparse usage of large on-die Touchstone files?


 

Should we support on-die "Quadrants" (e.g. Banks, Interfaces)?


 

Should we support full on-die models?

 

 

Walter Katz

wkatz@xxxxxxxxxx

Phone 303.449-2308

Mobile 303.335-6156

 

Other related posts: