disagreement with LISO

[cwipf]

Here's a little test case where Elektrotickle seems not to reproduce LISO's output. Am I doing something wrong?

------- test.fil --------
op u1 ad829 u1plus u1minus u1out
r r1 120 u1minus gnd
r r2 2.2k u1minus u1out
r r3 50 u1out gnd
uinput u1plus 0
uoutput u1out:db:deg
freq log 1e2 10e6 1000
------- cut here -------

------- opamp.lib -------
name = ad829 # fast lownoise voltage-feedback, No ext. comp., measured at gain 1
a0 = 1e5 # 100V/mV
gbw = 85M
un = 1.7e-9
uc = 30 
in = 1.5e-12
ic = 100 # not given in datasheet
umax=12
imax=0.05
sr=260e6
zero = 19.3M # fitted from measurement
pole = 53.4M 5.1 # fitted from measurement
zero = 76.5M 4.7 # fitted from measurement
pole = 13M # fitted from measurement
delay = 1.9n # fitted from measurement
------- cut here -------

------- Elektrotickle input -------
>> test=load_liso(load_opamps('opamp.lib'), 'test.fil');
>> [f, sigAC, noiseAC] = test.tickle();
>> semilogx(f, db(sigAC(test.getVariableIndex('node', 'u1out'), :)))
------- cut here -------

LISO says:

Elektrotickle says:

[tobin]

Thanks for the test case. I'll take a look!