[SI-LIST] Re: IBIS Model Quality (or lack thereof)
- From: "Jeremy Plunkett" <jeremy@xxxxxxxxxxxxxxx>
- To: freeman@xxxxxxxxxxx
- Date: Tue, 26 Mar 2002 14:43:43 -0800
Hi Jim,
my understanding based on the hspice manual is that the _tune parameters =
are used to determine the turn-off time(VT curve) of the transistors =
being disabled at a given edge, based(proportionately) on the VT curve =
of the transistors that are being enabled on that edge. All of =
ServerWorks' IBIS models include turn-on and turn-off VT curves for each =
transistor bank (pullup or pulldown), so I would not expect the _tune =
parameter to have any effect as long as I am setting ramp_rwf and =
ramp_fwf equal to 2 so that all VT curves are used. If I don't set them =
to 2, I don't expect to see good correlation with the original spice =
netlist, so I don't usually try that, although I have once or twice.
This is just my own interpretation of the hspice manual; please let me =
know if you think that I should be using the _tune parameters for =
something. I would hesitate to test our models with them set to other =
than the default in any case because I expect(hope) that hspice's =
defaults are similar to the default usage of the IBIS data for other =
simulators, and I would like my hspice results to be consistent with =
other simulators.
regards,
Jeremy
-----Original Message-----
From: Jim Freeman [mailto:freeman@xxxxxxxxxxx]
Sent: Tuesday, March 26, 2002 2:12 PM
To: jeremy@xxxxxxxxxxxxxxx
Cc: twester@xxxxxxxxxxx; si-list@xxxxxxxxxxxxx
Subject: Re: [SI-LIST] Re: IBIS Model Quality (or lack thereof)
Hi Jeremy,
What values do you use for fwf_tune and rwf_tune in your simulations =
with
Hspice?
Thanks
Jim Freeman
Jeremy Plunkett wrote:
> Todd,
> At a minimum, ServerWorks' IBIS models are validated prior to release =
by:
>
> 1) Syntax check using Hyperlynx Visual IBIS editor.
> 2) Correlation to the original spice deck waveforms using the hspice B
> element. In the past, the degree of correlation has not been =
quantified
> because typically it is all-or-nothing; either the waveforms are a =
99%
> match point-for-point (inspected visually in Awaves), or there was =
some
> serious problem in the conversion that needs to be corrected. =
Occasional=3D
> ly,
> there are correlation issues caused by limitations of the IBIS format =
or =3D
> the
> extraction process; in these cases we determine a work-around such as
> changing which model data is included in which curves, adding a =
submodel,
> etc. When a standardized method of correlating the IBIS results to =
spice=3D
> is
> determined by the IBIS qualification committee I would be happy to =
includ=3D
> e
> it in our correlation process, provided it can be accomplished either =
usi=3D
> ng
> hspice MEASURE functions such as DERIVATIVE, ERR, etc, or using a =
softwar=3D
> e
> utility that can import the waveforms from a text file.
>
> After the initial release, our models are converted to various other =
form=3D
> ats
> in-house and by our customers, which can sometimes bring out =
correlation
> issues between different simulators or problems with the model =
conversion
> processes. Also, at some point after the initial release the models =
are
> correlated with silicon and corrected if necessary.
>
> Serverworks has the same situation that Adam Tambone mentioned in his
> posting, a large library of existing IBIS models(though not as large =
as
> Fairchild's), not all of which were generated or correlated to the =
same
> level of scrupulousness that we currently follow. In our case most of =
th=3D
> ese
> parts are no longer being used in new designs, so fixing problems in =
the
> IBIS models is a low priority, however we will address any issues with =
an=3D
> y
> of our models as customers request it.
>
> To any customers who may be reading this: feel free to address any =
feedba=3D
> ck
> regarding ServerWorks' IBIS models directly to me (preferably =
privately, =3D
> not
> to the whole SI list). However, please address requests to receive =
the
> models in the 1st place to our marketing dept.
>
> Jeremy
>
> |>--/\/\/--((((((((()--|>
>
> Jeremy Plunkett
> Signal Integrity Engineer
> ServerWorks Corp
> www.serverworks.com
>
> |>--/\/\/--((((((((()--|>
>
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx
> [mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Todd Westerhoff
> Sent: Monday, March 25, 2002 12:59 PM
> To: si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] IBIS Model Quality (or lack thereof)
>
> Hi all,
>
> We have been doing a lot of comparisons between IBIS and HSpice models
> lately, with a lot of less-than-satisfactory results.
>
> I have an open question for the semiconductor suppliers among us:
>
> What testing is performed after an IBIS model is created, to ensure =
that =3D
> the
> completed model behaves as expected? I'm interested in responses from
> vendors that actually SIMULATE the IBIS model before distributing it =
for
> customer use. It looks to me as though many of the models I've =
received
> were never used in a simulation before being released ... and, as a
> potential customer, I find that quite disheartening.
>
> So the question is - what processes are being used to test IBIS models =
an=3D
> d
> ensure they function properly before their release to customers?
>
> As always, replies both on and off the list are greatly appreciated.
>
> Todd.
>
> Todd Westerhoff
> Signal Integrity Engineer
> Hammerhead Networks
> 5 Federal Street - Billerica, MA - 01821
> email:twester@xxxxxxxxxxx - ph: 978-671-5084
> =
=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D=
3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D
> =
=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D=
3D=3D3D=3D3D=3D3D=3D3D
>
> "I=3D92m starting With the man in the mirror
> I=3D92m asking him to change his ways
> And no message could have been any clearer
> If you wanna make the world a better place
> Take a look at yourself, and then make a change"
>
> - "Man in the Mirror", Michael Jackson
>
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
> or to administer your membership from a web page, go to:
> //www.freelists.org/webpage/si-list
>
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
> List archives are viewable at:
> //www.freelists.org/archives/si-list
> or at our remote archives:
> http://groups.yahoo.com/group/si-list/messages
> Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
> or to administer your membership from a web page, go to:
> //www.freelists.org/webpage/si-list
>
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
> List archives are viewable at:
> //www.freelists.org/archives/si-list
> or at our remote archives:
> http://groups.yahoo.com/group/si-list/messages
> Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
