[ibis-macro] Re: Summary and recommendations regarding Analog Buffer Modeling discussion (Tstonefile, ...)
- From: "Walter Katz" <wkatz@xxxxxxxxxx>
- To: <Arpad_Muranyi@xxxxxxxxxx>, "'IBIS-ATM'" <ibis-macro@xxxxxxxxxxxxx>
- Date: Wed, 07 Dec 2011 15:51:50 -0500 (EST)
Arpad,
Comments in-line.
Walter
From: ibis-macro-bounce@xxxxxxxxxxxxx
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Muranyi, Arpad
Sent: Wednesday, December 07, 2011 3:02 PM
To: 'IBIS-ATM'
Subject: [ibis-macro] Re: Summary and recommendations regarding Analog
Buffer Modeling discussion (Tstonefile, ...)
Walter,
I am not sure what you are trying to say with this most
recent email. Please explain.
Regarding the earlier email, I would like to ask you some
questions about the idea of using those new "Language"
entries for [External Model]. It seems that when the
language is set to Touchstone, Generic_Tx or Generic_Rx,
you would basically reference the corresponding assumed
circuit topology and describe what those circuits are
somewhere in the specification. There would be no additional
circuit files or netlists involved. The parameters you
list under each of these languages would correspond to
the predefined circuit topology's elements. Correct?
WMK> Correct
So here is the detail question: Currently anything inside
an [External Model] is connected through the seven reserved
ports (nodes) of a normal [Model]. If there is an analog
model insider [External Model] we need to go through the
D_to_A and A_to_D converters, because it is assumed in
IBIS that the tool's internal stimulus is digital.
WMK> This is why I always objected to using [External Model] for AMI
analog models.
The input and outputs of AMI buffer models are not digital, they are
analog.
How do you envision this connection to your predefined
circuits? Are you thinking of going through those seven
reserved nodes as normal, or do you envision something
different?
WMK> The connection to the Generic Tx and Generic Rx are defined by the
circuits that are described "somewhere in the specification".
The reason I am asking is because I wonder where and how
those Vol and Voh parameters are going to be applied.
Are you going to control those converters with them?
Or are these stimulus sources included inside your
predefined circuits? In other words, I wonder where the
boundaries are between the tool's stimulus generator and
the predefined circuit? (This is basically a repeat of
the questions Scott asked about these analog models at
the beginning of this year).
WMK> BIRD 122 clearly defines how Vol, Voh and Trf are applied to the Tx
Tstonefile and Generic Rx models.
Thanks,
Arpad
===========================================================
From: ibis-macro-bounce@xxxxxxxxxxxxx
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Walter Katz
Sent: Wednesday, December 07, 2011 1:32 PM
To: 'IBIS-ATM'
Subject: [ibis-macro] Re: Summary and recomendations regarding Analog
Buffer Modeling discussion (Tstonefile, ...)
All,
I clarified the statement about "Equivalent-Circuit Rx and Tx defined in
BIRD 122" to include" in the models AMI operating range".
Walter
From: ibis-macro-bounce@xxxxxxxxxxxxx
[mailto:ibis-macro-bounce@xxxxxxxxxxxxx] On Behalf Of Walter Katz
Sent: Wednesday, December 07, 2011 12:18 PM
To: IBIS-ATM
Subject: [ibis-macro] Summary and recomendations regarding Analog Buffer
Modeling discussion (Tstonefile, ...)
All,
First I will review various features of BIRDS 116, 122, and 144, and then
make several recommendations.
For [Model]s that have an AMI model
Tstonefile
BIRD 116 requires an ISS subckt file containing an S element instance, and
requires parsing a number of lines in the [External Model] section to
determine what is defined the BIRD 122 AMI parameters Tstonefile,
Nodemap, Voh, Vol, Trf.
Equivalent-Circuit Rx and Tx defined in BIRD 122
The Equivalent-Circuit Rx and Tx defined in BIRD 122 accurately represent
the differential response in the models AMI operating range of legacy IBIS
models that have a constant differential impedance and do not have VT
curves (i.e. Use dV/dt_r and dV/dt_f instead of VT curves). This includes
Series Pin and Termination models. (The Equivalent-Circuits will need to
be enhanced to include some inductors).
Differential ISS subckts.
Both BIRD 122 and 116 enable a model to define an File and Subckt for a
Differential ISS subckt.
For [Model]s that do not have an AMI model
BIRD 116 allows the user to use an ISS subckt to represent the Analog
Model
Recommendations
Enhance BIRD 116 to allow the following Languages, and keywords for each
language (keyword names and syntax TDB)
Touchstone
Tstonefile
Nodemap
Vol
Voh
Trf
Generic_Tx
Vol
Voh
Trf
Rs
Rd
Rt
Cd
Cc
Rac
Cac
Ls
Lt
Ld
Generic_Rx
Rd
Rt
Cd
Cc
Rac
Cac
Lt
Ld
ISS
As defined in BIRD 116
BSS
This is same as ISS, except the subckt is compliant with IBIS Buffer Spice
Subckt (IBIS BSS). IBIS BSS is a superset of IBIS ISS. IBIS BSS contains
additional elements, such as PWL controlled sources, event detectors, and
timers. IBIS BSS subckt can be written to be equivalent to a compliant
legacy "B" element.
Approve both BIRD 122 and 116 (or combine them into a single BIRD, so that
AMI model developers can put Tstonefile, Equivalent-Circuit Rx and Tx, and
Differential ISS subckts in either the .ami file and or the .ibs file
Reject BIRD 144
Walter
Walter Katz
wkatz@xxxxxxxxxx
Phone 303.449-2308
Mobile 720.333-1107
Other related posts:
