Re: WaVEFORMS FOR BIRD12.1

Will,

I see the confusion. What I meant was that for Case (2), you would
provide 4 waveforms for each of the R_fixture and V_fixture combinations
listed. In Case (3), 6 waveforms would be provided. Jon's document
showed six waveforms, all of which could be included in the IBIS [Model]
description. I have no objection to having as much waveform characterization
information as possible provided. It becomes a simulator vendor's problem
to figure out how best to use this information (or to choose what information
not to use). I raised the question to see if there were any preferred
combinations, and any preferred ways of dealing with multiple sets of data.

Bob
> Bob,

> If I understand your question, for the Vcc/2 waveform,
> wouldn't (4) be the only valid option? Option (2) would
> also be correct if the load were 100 ohms each, for a total
> Thevenin equivalent of 50 ohms. Or am I out in left field?

> Will

> To Jon Powell, Arpad Muranyi, and IBIS Committee:

> This is an open implementation preference question based on the waveforms
> you have sent. In particular, refer to Jon Powell's "Synchronizing
> Measure AC Waveforms ... " document recently sent which illustrates
> the synchronization of rising and falling waveforms measured into
> 50 Ohm loads at 0, 2.5 and 5V.

> How do you "blend" certain data combinations that would be allowed via the
> BIRD12 [Rising waveform] and [Falling waveform] mechanisms? And which
> form would you prefer seeing?

> Cases --> (1) (2) (3) (4)

> Low_To_High:
> 50 Ohm to Vcc * *
> 50 Ohm to Vcc/2 * *
> 50 Ohm to GND * * *

> High_To_Low:
> 50 Ohm to Vcc * * *
> 50 Ohm to Vcc/2 * *
> 50 Ohm to GND * *

> Note, Case (1) corresponds to the [Ramp] specification conditions in IBIS
> Version 1.1.

> Bob Ross
> Interconnectix, Inc.
Received on Wed May 11 09:32:35 1994