S2IBIS update

Sorry if this is a repeat message but our earlier
posting did not seem to turn up.

7/20/95

Here is an update on our (North Carolina State University) progress on the
next spice-to-ibis converter (s2ibis).

We have been able to undertake this project thanks to the generous
support of Cadence.

The basic strategy is to parse an s2ibis input file, set up internal data
structures completely (with the exception of the data tables for VI and
tV curves) and then initiate analyses with a flavor of Spice. These
analyses are parsed and the results used to complete the internal ibis
data structures. The *.ibs file is then generated from the completed
data structure. This is a departure from the way version 1 of s2ibis
works--it was essentially an interpreter that fired off analyses as the
input file was parsed. The new scheme does require more memory than
s2ibis v1.1, and the command file can be more complex, but the new
grammar seems to be the only reasonable way to embody the complexity
allowed by the IBIS v2.1 specification, to provide meaningful message
reporting, and to provide for relatively painless future expansion.

We estimate that we have put 7 man-weeks into the coding so far. We
have completed the output code, parts of the analysis code (thanks to
Steve Lipa for his trailblazing efforts here), and are currently
addressing some initial deficiencies in our command grammar.

We expect to have a beta version of the software (including
documentation and a cut-and-paste command file template) within four
weeks. The prototype will allow for all features contained in the IBIS
v2.1 specification, with the exception of the Package Modeling
capabilities. While our converter will handle the [Package Model]
keyword correctly, it will not handle any of the information that can be
contained in a .pkg file.

The converter currently is compatible with the following flavors of
Spice: HSpice, PSpice, Berkeley Spice 2 and Berkeley Spice 3. Support
for Cadence's Spectre is conditioned on availability of the software for
testing; this issue is currently being resolved.

Michael Steer
Alan Glaser
Received on Fri Jul 21 07:49:54 1995