I also remember looking at this, several years ago.
?
This should be fixed. For practical components, the common mode I/P impedance is always higher than the differential I/P impedance. So, I'd say this was a bug.
Is it possible that this behavior was intentional?? It seems unlikely, but ....
?
Might it have been done this way to be compatible with op-amp behavioral models in certain other SPICE simulators?? Does PSpice (for example) have a "universal" op-amp behavioral model too?
?
Would fixing this now "break" existing simulations?
?
Would you trust handing it over to someone to fix it, and having them not also "fix" other things that did not need fixing?? I do not have much confidence in that right now.? They break things that did not need to be broken - even if they later unfix the things they broke.? They do not seem to have the wisdom and maturity to leave some things alone.
?
Andy
?
