RE: [SI-LIST] : ibis or spice?

About this list Date view Thread view Subject view Author view

From: Moran, Brian P (brian.p.moran@intel.com)
Date: Thu May 03 2001 - 09:33:27 PDT


John,

In most cases where I have cited a perceived deficiency in IBIS, an IBIS
advocate has usually been able to point out a rational to dispell my
concern, or a mechanism for overcoming the problem within the framework of
IBIS 3.2. So I have slowly become a convert to the IBIS methodology.
However, two problematic areas that most IBIS advocates concede are in
accuartely modelling power and ground distribution between the pin and the
on-die buffer, and modelling current sharing differential output buffers.
Assuming semi-ideal power and ground and single ended operation, the IBIS
methodology yeilds equivalent results in much less time. In fact in many
cases it can be used to simulate differential buffers as well. I have
generally found the larger issue to be the accuracy of the curve data and
the assumptions made in generating the data and the model. Its possible to
create poor models in either case. It should also be said that it is hard
to generalize about such things. Each case has to be looked at in terms of
the simulation scope & objectives, the tool environment, and the models
available. If your focusing on a single interconnect and have transistor
level power delivery and buffer models then you'll get the best results in
HSPICE, but if you need to do board level SI screening and timing
verification, get the most current IBIS models avaiable, and run IBIS
simulations. Most chips and boards we do have a mixture of simulation
methodologies used, based on the nuances of the various interfaces.

Brian P. Moran
Signal Integrity Engineer
Intel Corporation
brian.p.moran@intel.com

-----Original Message-----
From: Bo [mailto:bo_pfc@yahoo.com]
Sent: Wednesday, May 02, 2001 6:20 PM
To: John Zhang; si-list@silab.eng.sun.com
Subject: Re: [SI-LIST] : ibis or spice?

Here is a political answer to your question:

The one that gives you results faster with more accuracy... If you are sure
that your IBIS model will give you as accurate results as SPICE you'll be
fool
not to use it. IBIS is in most cases way faster than SPICE.

Bo

--- John Zhang <ibis_or_spice@yahoo.com> wrote:
> Question for model users:
>
> What is the preferred model and why? IBIS or Spice?
>
> I know people who hate IBIS and people who like IBIS.
> So far, I've yet to meet someone that hates Spice.
>
> Any comments as to which models you would prefer?
>
> Thanks.
>
>
>
> __________________________________________________
> Do You Yahoo!?
> Yahoo! Auctions - buy the things you want at great prices
> http://auctions.yahoo.com/
>
> **** To unsubscribe from si-list or si-list-digest: send e-mail to
> majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
> si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
> si-list archives are accessible at http://www.qsl.net/wb6tpu
> ****
>

__________________________________________________
Do You Yahoo!?
Yahoo! Auctions - buy the things you want at great prices
http://auctions.yahoo.com/

**** To unsubscribe from si-list or si-list-digest: send e-mail to
majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****

**** To unsubscribe from si-list or si-list-digest: send e-mail to
majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Jun 21 2001 - 10:11:49 PDT