Elasto-plastic bending - elements, accuracy

This is an example of what I’m obtaining.

Based on Professor M.Kraska tensile test of a wire I have impose a displacement in the longitudinal direction of 0.3mm.

Depending on the shape of the amplitude card the result is completely different.

Stretch linear A=0.3*t for t [0,1] , uniform Stress and uniform PEEQ. No necking.

Stretch applied with a half sinus A=0.3*Sin(pi()*t/2) for t [0,1] asymmetric Stress and PEEQ values concentrated in one side.

Stretch applied with a third function I’m still working with for t [0,1] symmetric Stress and PEEQ values around the center.

Kind of weird all togheter ¿isn’t it¿.

Is there an explanation for this?.
Someone has suggest there is some imperfection needed to properly initiate the necking but this can’t be done on a real model as one doesn’t know where the plastic behaviour will start.