Re: [IBIS] Question regarding V-t tables

Arpad:

One way to mimic test-case data-dependent simulations in IBIS
(different waveforms for the same fixture load) is to use this
trick:

Store 1-1/2 cycles of a data-dependent "rising edge"
data for the [Rising Waveform] and the corresponding 1-1/2
cycles of the corresponding data-dependent "falling edge" for
the [Falling Waveform]. That way the two different waveshapes
are captured in one model for a test-case setup run at 1/2
speed.

Bob

Muranyi, Arpad wrote:

> Thanks Chris, Bob, and Andy for your responses. Let me give a simple
> example for the reason I was asking this.
>
> I made an IBIS model for a differential buffer. Two V-t curves were
> generated for it, one rising and one falling waveform. One of them
> was saved from the non-inverting pad, and the other from the inverting
> pad. The input was stimulated with only one edge. The IBIS model
> correlates very well on every other transition, but not so well on
> the rest of the transitions, see:
>
> http://www.eda.org/pub/ibis/summits/oct03/muranyi.pdf on page 34.
>
> I didn't verify this yet, but I suspect that if I would have generated
> two more waveform tables with a stimulus going the opposite way and used
> those waveforms for the transitions that don't match so well, I would
> have gotten all transitions match well.
>
> In this particular case I could probably do it with regular IBIS
> syntax if I made two ***different*** [Model]s with the two sets of
> waveforms
> (one for the "P" pad and the other for the "N" pad) and associated them
> with the normal [Diff Pin] keyword.
>
> However, there may be other, more complex situations when this may not
> be possible. In those cases I cold envision a *-AMS solution which
> still
> uses regular IBIS data, but adds a waveform selector logic that is based
> on a set of logic equations which can detect data patterns and select
> the
> corresponding V-t tables on the fly for a given transition or pattern
> from
> the otherwise normal IBIS [Model].
>
> This was the reason for my question. How could I put multiple waveform
> tables in a [Model] with the same Rfixture and Vfixture parameters, and
> tell the simulation engine which one to use depending on the input
> stimulus (rise or fall), or depending on the data pattern on the input
> (0101010, or 00001111, etc)? Please note that the I-V and V-t curve
> processing algorithm would be the same as before otherwise, i.e. basic
> IBIS algorithms would still work. This is just like a [Model Selector]
> like Bob suggested, except that it does it on the fly, and could be
> controlled by more complicated data pattern definitions.
>
> I know, this could be done completely in *-AMS without touching the
> legacy portions of the IBIS spec, so I am not pushing for this too hard.
> I just wanted to gather some opinions to see if it would be worth
> writing
> a BIRD on this, or forget it and do it strictly in *-AMS.
>
> Any (new) feedback would be appreciated.
>
> Thanks,
>
> Arpad
> ========================================================================
>
>
> -----Original Message-----
> From: owner-ibis@eda.org [mailto:owner-ibis@eda.org] On Behalf Of Andrew
> Ingraham
> Sent: Wednesday, September 08, 2004 9:35 PM
> To: ibis@eda.org
> Subject: Re: [IBIS] Question regarding V-t tables
>
> Arpad,
>
> Rather than the question "what would the simulator do with it," I find
> the more important question is, "can IBIS even model a device that has
> data pattern dependent behavior?"
>
> I think the answer currently is no.
>
> So one wonders why someone would even attempt to develop an IBIS file
> for such a device. I think it's simply not appropriate.
>
> Wouldn't this be like trying to represent an input-hold circuit by just
> having multiple values in the same V-I table with the same voltages?
>
> Regards,
> Andy
>

-- 
Bob Ross
Teraspeed Consulting Group LLC     Teraspeed Labs
121 North River Drive              13610 SW Harness Lane
Narragansett, RI 02882             Beaverton, OR 97008
503-750-6481                       503-430-1065
http://www.teraspeed.com           503-246-8048 Direct
bob@teraspeed.com
-----------------------------------------------------------------
|For help or to subscribe/unsubscribe, email majordomo@eda.org
|with 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 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