Subject: RE: [IBIS-Users] C-comp question
From: Muranyi, Arpad (arpad.muranyi@intel.com)
Date: Wed Feb 19 2003 - 16:37:57 PST
Garry,
These observations are reasonable. We had long debates over this
in the discussions when we wrote the spec. However, another way
of looking at the problem is this: Is your slow case the worst
case and the fast case the best case? Some may say yes, others
may say no, and they are both correct...
A receiver having a small C could give you the fastest edge rates
at the receiver, but it could also allow the signal to overshoot,
and consequently ring like crazy. A large C, on the other hand,
could slow your edge rate down, but eliminate overshoot and ringing
too. Which one do you consider best/worst or fast/slow case?
If your ringing is so bad that it reenters the threshold region,
this fast case could give you the longest timings, because you
will have to wait for the ringing to settle.
If you add resonance conditions, it may not even be the min or
max capacitance value that overshoots and rings back the most,
it could be anywhere in-between (or outside).
Examples like this were the reason that we decided not to associate
typ/min/max with any particular fast/slow, best/worst qualifiers.
To cover all your bases, you should really simulate all possible
combinations of the non-tracking (independent) parameters in the
IBIS model. However, this may take forever, and not everyone may
be interested to do it all the time.
For these reasons (and perhaps others) the selection of these
parameters is really best placed into the GUI of the tool.
Arpad
==================================================================
-----Original Message-----
From: garry.felker@conexant.com [mailto:garry.felker@conexant.com]
Sent: Wednesday, February 19, 2003 4:18 PM
To: Muranyi, Arpad
Subject: RE: [IBIS-Users] C-comp question
Further comments I received...
Garry
See IBIS 3.2 spec page 76/77. The exact wording is "The conservative
approach
to using IBIS data will associate large C_comp values with slow, weak
models, and the small C_comp values
with fast, strong models."
The slow model is what we call min. So the large value goes to min column.
The whole thing will not make sense until you think of these as corner, not
absolute values.
If you read pp 76, that is the intention. Indeed the recommended approach
regarding C_comp is very conservative. It will only make the slower corner
slower than what it really is and fast corner faster than what it really
is.
Comments?
----- Forwarded by Garry L Felker/USA/Conexant on 02/19/2003 04:14 PM -----
|------------------------------------------------------------------
|For help or to subscribe/unsubscribe, email majordomo@eda.org
|with just the appropriate command message(s) in the body:
|
| help
| subscribe ibis <optional e-mail address, if different>
| subscribe ibis-users <optional e-mail address, if different>
| unsubscribe ibis <optional e-mail address, if different>
| unsubscribe ibis-users <optional e-mail address, if different>
|
|or email a written request to ibis-request@eda.org.
|
|IBIS reflector archives exist under:
|
| http://www.eda.org/pub/ibis/email_archive/ Recent
| http://www.eda.org/pub/ibis/users_archive/ Recent
| http://www.eda.org/pub/ibis/email/ E-mail since 1993
This archive was generated by hypermail 2b28 : Wed Feb 19 2003 - 17:03:54 PST