Hi,
From this link:
it was recommended to use shooting for circuitry that has sharp transitions, but hb for circuitry that has s-parameters.
My circuit has a series of buffers (sharp edge) and also s-parameter to reflect some routing so I tried both shooting and harmonic balance. I did not notice significant difference in pss sim itself, but hb take much less time for noise analysis(i use the same no. of harmonics). Is there a reason why?
The other observation, is that noise analysis result on the nodes along the chain of buffers are the same between hb and shooting, but for the nodes before and after sp file, they are quite different. Since hb is recommended for sp model, does that mean we should trust result from hb instead of shooting? in other words, it's more an accuracy concern than efficiency ?
thanks,
Kevin