Re: [Gnucap-devel] bug in pulse

[Felix Salfelder]

Hi Al.

On Sat, Jan 11, 2014 at 03:41:00PM -0500, al davis wrote:
> Spice compatibility isn't the issue here.  Spice does not allow 
> 0 rise or fall time.  It substitutes the user time step from the 
> tran command.

all the better.

i'm still convinced that the pulse should evaluate to iv at delay. or
more general it should evaluate to the value before a transition until
after the transition has started. (your patch does not affect this).

here's another example.

"""
spice
V1 1 0 pulse iv=0 pv=1

.print dc v(nodes)
.dc
.print tran v(nodes)
.tran 0 1m 1m trace=a
.end
"""

imo the initial DC should yield v(1)=0. consider an additional
resistor(1,2) and a capacitor(0,2). only an empty capacitor at time=0
makes sense to me. now if spice does replace rise with something
positive, it should also end up with an empty cap in .dc.

thanks
felix