Re: clarification on IBIS min and max values

On Tue, 14 Nov 2000, Ingraham, Andrew wrote:
> A simulator does not need to set all parameters to their "min"
> values", or "max" values, simultaneously. It can mix-and-match any
> way it sees fit.

Actually, although this was the original concept, the data really
must be looked at as 3 data sets, but there is some crossing between
min and max. Some simulators may max-and-match, so they can do
what-if calculations, but in doing so they are using inconsistent
data.

Consider ....

1. If you take one pullup/pulldown, with a different set of
waveforms, the end points will not match.

2. If you take one voltage_range or pullup_reference with a different
set of waveforms, the end points will not match.

3. Rise/fall times are dependent on c_comp, and pullup/pulldown.

4. The only place temperature is used is in the transit time
calculation which describes the nonlinear capacitance of the clamps.
This is part of the load, so if you mismatch, waveforms could be off.

So, consider them to be data sets, but unfortunately, you can't just
take a column. It is twisted a bit.

al.
 
Received on Wed Nov 15 01:33:38 2000