IBIS to SPICE

Hi D.C., all

D.C. wrote:
>Another aspect: as I observed some time ago, for those of us using
>SPICE sims to generate IBIS models an IBIS->SPICE translator would
>also allow us to QA the IBIS models by translation back with
>subsequent correlation.

 This would be a great improvement over not testing an IBIS model.
I would hope there are not any companies out there that ship IBIS models
without testing them in some way. If there are any IC companies doing this
please contact me and I will do my best to help you get setup with a simulator.

 If you do use a SPICE to IBIS translator to check the IBIS model it may or may
not prove out your model. It all depends on the quality of the IBIS translator. This
is not a trival task. I am very concerned about a quick and dirty IBIS to SPICE
translator that only works for a very limited set of cases. This could cause significant
problems for users that then blame the IBIS model when the problem is the translator.
IBIS is a very powerful standard but it does take a significant amount of work to do it
right. Anyone that creates a translator in a few weeks has missed most of the really
hard to find triva that is need to make things like LVDS or ECL or 2 stage outputs work
correctly to mention just a few items.

 I still believe the best solution is to beat on the remaining SPICE companies that haven't
added IBIS modeling. Why aren't more of their users demanding it be added to SPICE?
It should be only slightly more difficult to add it to SPICE then to create a really good
translator. This way you know the vendors will take over support of the simulation and make
sure it works. With a translator created by someone and put in the public domain you
mayl have no idea what it doesn't do correctly and no where to turn to get it fixed.

-------------------------------------------------------------------------
Have a great day...
Kellee Crisafulli at HyperLynx
kellee@hyperlynx.com http://www.hyperlynx.com
-------------------------------------------------------------------------
 
Received on Mon Oct 27 11:15:31 1997