[uunet!icx.com!bob: Re: IBIS Spec]

Bob and Paul et. al.

The "Timing Correlation" remark here, though indeed a function of
loading, is actually to allow for the detection of rise-time-controlled
component drivers. In these cases the Pull-up and pull-downs turn on and off
with delay (So the pull-up would turn off before the pull-down turns on etc).
There is no way to test for this in the parser. There are perhaps so complicated
"suspicious data" tests you could do but I don't think I would recommend it at
this point.

jonp

Return-Path: <uunet!icx.com!bob>
Date: Mon, 3 Oct 94 17:06:06 PDT
From: uunet!icx.com!bob ( Bob Ross)
To: uunet!compuserve.com!73053.721, uunet!vhdl.org!ibis
Subject: Re: IBIS Spec

Paul,

I found your timing question: I believe that this statement should be for
developer guidance only, but NOT for IBIS_CHK. The intent is for the same
stimulus reference time (or input) to be used for all rising waveforms, and
similarly, the same stimulus reference time (which can differ from the rising
waveform) or input be used for all falling waveforms.

Thus all variations are due only to to loading changes. The simulator can
assume common input reference times for possible interpolation algorithms.
There is no way within IBIS to specify the input waveforms for output
responses, so constraining them to a common reference time is one way
of eliminating the effects of input waveform variations.

Bob Ross
Interconnectix, Inc.

> Bob,

> I had a question as to some usage rules for rising and falling waveforms.
> Under [Rising Waveform] and [Falling Waveform] keywords, the 3rd paragraph
> states "If more than one rising or falling edge waveform table is present, then
> the data in each of the respective tables must be time correlated. In other
> words, the rising (falling) edge data in each of the rising (falling) edge waveform
> tables ust be entered with respect to a common reference point on the input
> stimulus waveform."

> I don't understand what this is trying to say. Is this something the parser can
> check for, or is it just something that the developers of IBIS files need to keep
> in mind?

> Paul Munsey
Received on Tue Oct 4 08:43:29 1994