Re: [Help-gnucap] Possibly a bug?

[al davis]

On Wednesday 22 May 2013, Orestes Mas wrote:
> If soo, why not do an OP analysis automatically before every
> AC one? If  I remember correctly, this is precisely what
> spice does...

In this case .. there is a bug that causes a divide by zero on a 
zero-ohm resistor, with a simple fix.  (Does this also happen 
with a zero-Henry inductor? .. need to check)

To answer your question of why not automatically do an op first 
...   It's more flexible this way.  Instead of forcing a nominal 
static condition, gnucap uses the most recent point from the 
most recent dc, op, or tran, which lets you do AC analysis at 
non-static points.

For a simple example, consider a class-B amplifier, underbiased 
a little so there is a crossover notch.  The usual op would set 
the operating point in the middle of the notch, so the AC 
analysis would be useless, other than to show how bad the notch 
is.  

Run a transient analysis to get it out of the notch, stop there, 
do AC there.  See how it varies with signal swing.  Try it with 
different frequencies in the transient analysis.  The results 
can be very different from any static guess.