Re: paramset resolution

Geoffrey.Coram wrote:
> Kevin Cameron wrote:
>   
>> Geoffrey.Coram wrote:
>>     
>>> I think that's right: if you didn't specify all the connections,
>>> then you're content with a less-accurate -- and presumably faster! --
>>> model.
>>>
>>>       
>> I prefer to have things "fail safe", i.e. default to the behavior that's
>> least likely to be wrong (use the accurate model).
>>
>> Bad silicon is expensive. As a user you don't want tools making the
>> wrong assumptions for you.
>>     
>>> Eg, if you don't specify the substrate connection of a bipolar,
>>> you're content with a 3-terminal model.
>>>
>>>       
>
> If you didn't specify the substrate connection, then it's a
> bad idea to select a model with 4 terminals -- what happens
> with the 4th terminal?  I would say it's correct to choose
> the 3-terminal model, if one is available.  If you want to
> make sure the most accurate models are used, then there
> shouldn't be a 3-terminal model available!
>
> -Geoffrey
Are you saying unconnected terminals should be disallowed? :-)

I don't see anything wrong with plugging a 3 terminal model into a 4 
terminal socket (or vice versa) if the models are designed for that, 
it's just that given the tools generally have no understanding of what 
the specific terminals do there needs to be some syntax/semantics that 
says what should happen explicitly.

This touches on the issue of power distribution since presumably if you 
want to plug a 4-terminal model into a 3-terminal socket you would like 
to be able to tie the substrate off to vdd/gnd.

Kev.