Re: [Help-gnucap] Time step control

[al davis]

On Monday 03 December 2007, a r wrote:
> BTW, just an observation.
>
> Internal bsim models and "exact spice models" are quite
> different. This is especially visible when simulating a ring
> oscillator. I had a feeling before that internal models give
> somewhat overoptimistic results (transistors were faster that
> they usually are at 0.25u).

I am not surprised.  The DC part is ok.  I had a lot of trouble 
with the dynamic part.  This experience was a big factor in 
convincing me that the model strategy needed to change.  I 
don't want to port models.  Gnucap needs to be able to use 
models developed by others with no changes.

In the future there will be no built-in models.  That's what 
plugins are for.