Re: kf in mos1 model

On Tue, May 06, 2014 at 07:54:53AM +0200, Felix Salfelder wrote:
> On Mon, May 05, 2014 at 11:01:13PM -0400, al davis wrote:
> > My guess is that you have changed one of the headers, changing
> > the size of something, you are doing a partial compile (the
> > usual "make") and something is out of sync.  Perhaps you are
> > building a plugin, then plugging it to a version of gnucap that
> > had different headers.
> 
> everything i changed is the traces... after i discovered the issue.

Hi Al.

for the record, tests/==out/.diff indicates that i did not trigger the
issue myself. there are multiple chunks similar to

[..]
--- ==/charge-cons.ckt.out 2010-07-08 04:08:30.000000000 -0400
+++ 130913/charge-cons.ckt.out   2013-09-14 01:57:00.000000000 -0400
@@ -514,8 +514,8 @@
 M1 ( 0 2 0 0 )  m  l=50.u w=50.u nrd=1. nrs=1.
 .model m nmos2 ( level=2 kp=20.u nfs=0. neff=1. ucrit=49.K uexp=0.1 delta=0.
 + phi=0.6 tox=110.n nsub=2.2E+15 nss=32.G xj=2.95u uo=575. tpg=1 tnom=NA( 27.)
-+ fc=0.5 pb=0.8 cj=NA( 0.) mj=0.5 cjsw=0. mjsw=0.33 is=10.f rsh=0. cbd=4.5f
-+ cbs=4.5f cgso=1.5n cgdo=1.5n cgbo=0. ld=2.4485u)
++ fc=0.5 pb=0.8 cj=NA( 0.) mj=0.5 cjsw=0. mjsw=0.33 kf=NA( 0.) af=NA( 0.)
++ is=10.f rsh=0. cbd=4.5f cbs=4.5f cgso=1.5n cgdo=1.5n cgbo=0. ld=2.4485u)
 Gnucap   System status
 iterations: op=3, dc=0, tran=2589, fourier=0, total=3099
 transient timesteps: accepted=507, rejected=0, total=507
[..]

apparently, something causes kf and af to misbehave. which compilers did
you use?

thanks
felix