Re: IBIS BIRD64 - Package Model Selector

On Tue, 26 Oct 1999, micohen@us.ibm.com wrote:
> However, I do think that for the corner situations (TYP, MIN, MAX), we need
> to have a different keyword or structure. Or do we? I am looking for some
> discussion from the IBIS community on this matter. Does it make sense to
> use a "typical" package model for all three simulations (MIN, TYP, MAX)?
> Does it make sense to use a MIN package model in a MAX simulation?

Lots of combinations of typ/min/max make sense. I suppose even a completely
arbitrary mix makes sense. A simulator could do this now, without any changes
to the spec. Current simulators are not exploiting nearly the capability that
the existing spec allows. Desiging a reasonable user interface would be a real
challenge. Let's leave the spec alone, and let the simulator makers figure out
how to do it.

I am confused by the comment about an "alternate syntax".

It is not clear from the spec how this is placed within the file. Logically,
it can only go at top level. I suppose this would be implicitly stated by the
position in the spec.
Received on Tue Oct 26 23:03:37 1999