[ibis-macro] Re: Some observations on IBIS-Macro

  • From: "Muranyi, Arpad" <arpad.muranyi@xxxxxxxxx>
  • To: "IBIS Macro" <ibis-macro@xxxxxxxxxxxxx>
  • Date: Wed, 1 Mar 2006 13:08:32 -0800

A thought that I did not tell you in my previous comments
to jumped at me as I was reading this version of your
It seems that you expect the model maker to convert the
Vt curves to Kt, including the C_comp compensation part
of the process.  Seeing all the bad IBIS models out there,
and the kind of mistakes most of them have, I am afraid
that this task will go above the head of a lot of people,
and either turn them away from making models, or result
in terribly bad models.  We would either have to provide
a "canned" solution i.e. some s2ibis-like software that
does this well, or make the macro model more intelligent
so that it can take the Vt curves directly.
Add to the picture the complexity of characterizing
multi-tap drivers to the modeling process, we are already
pushing the IQ requirements for model makers quite a bit...


From: ibis-macro-bounce@xxxxxxxxxxxxx [mailto:ibis-macro-bounce@xxxxxxxxxxxxx] 
On Behalf Of Walter Katz
Sent: Wednesday, March 01, 2006 12:35 PM
To: IBIS Macro
Subject: [ibis-macro] Some observations on IBIS-Macro

I put some observations down about my experiences so far with applying 
IBIS-Macro concepts in the enclosed document. I have reviewed it with Bob, and 
Arpad, and incorporated many (but not all) of their comments. Bob and Arpad 
agree with many of my points, but disagree on some also.
On Monday afternoon I will email to the group all of the comments I will have 
received by then.

Other related posts: