I used ideal gms(voltage controled current source) build an ideal OTA with nondominant pole around 5-6times UBW, and compared its settling behavior with an real OTA that has identical gm and non dominant pole, the result from ideal OTA failed to repeat the same settling behavior of the real OTA. the real one don;t have overshoot, but the ideal one has 1% overshoot, anybody tell me why?