[SI-LIST] Re: Macromodeling

Kumar,

I hate to say this, but to me it seems that you are contradicting
yourself.  On one hand you say that you like SPICE because of its
simplicity and limitations, but on the other hand you say you wish
it had a "generalized equation element".

Isn't this equation element about the same what we have in AMS?
I know AMS can do more, but the differences are diminishing.
Since I started to use HSPICE, they added a lot of programming
language style capabilities too, such as "IF/ELSE" and similar
things.

One thing that I have a hard time to envision is that SPICE would
ever have mixed signal capabilities.  To me this is quite a big
feature, since the digital equations are event driven.  What that
means is that I can write really fast logic equations which only
execute when a value changes somewhere (I define it where), and
do not have to be evaluated for every single analog time step.  If
you added such things to your equation element, you would have AMS.

By the way, the complexity of AMS should not be a deterrent.  You
can use and think of AMS models the same exact way as it is done
by SPICE.  If it frightens you, you don't have to open the guts of
a model, you can instantiate it just like any SPICE element.  The
beauty to me is that if I have a need to develop my own "element"
I can do so.  I don't have to negotiate with my favorite tool
vendor to implement it for me, and wait months or years for the
implementation.  For example, Don and I have talked to the president
of Meta Software in the early 90-s to give us an even triggerable
PWL element in HSPICE to make it easier to write behavioral IBIS-like
buffer models.  (You have this in your PWL sources already, so you
know what a good thing this is).  HSPICE still doesn't have this
capability after 12-13 years and a few takeovers...

Arpad
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

-----Original Message-----
From: C. Kumar [mailto:kumarchi@xxxxxxxxx]=20
Sent: Thursday, March 24, 2005 1:54 PM
To: Muranyi, Arpad; si-list@xxxxxxxxxxxxx; owner-ibis@xxxxxxx
Subject: Re: [SI-LIST] Re: Macromodeling

actually my thinking is somewhat at varience.

1. I believe that the most attractive language is one
which  should include powerful but minimal set of
syntax. From my limited knowledge of  ams I tend to
view it as unduly complex for the task.  I am
concerned that it will actually stand in the way of
compact modeling and slow down real advnaces. I fear
of "natural" tendency to bloat the models with all
kinds of unnecessary overheads as in the inclusion  of
logic preprocessing circuits.=20

2. Spice is actually a very powerful language
precisely because of its limited constructs. It
provides connectivity information in the most basic,
abstract but elegant manner. That is why it is not a
huge task to go from a circuit schematic to a spice
description

3. The following fundamental additons will make it
unbeatable.
  a. arbitrary locally defined parameters and
parameter passing - many commercial simulators already
have this

   b. generalized equation element of the form

       f(v,i, .....parametrs) =3D 0;

      where v and i are voltage and current vectors.
    all the spice elements including  EFGH are a
variation of this general variety.=20

4. as we go into an era with even more serious analog
models (read network equation with node conservation
which spice is all about), I will be loathe to take on
another curve ball, which pretends to be high level
but not really but at the same time  too complex for
network description.  Such a development can only=20
obscure and deflect from real advances.
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field

or to administer your membership from a web page, go to:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List FAQ wiki page is located at:
                http://si-list.org/wiki/wiki.pl?Si-List_FAQ

List technical documents are available at:
                http://www.si-list.org

List archives are viewable at:     
                http://www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu

Other related posts: