A shock repsonse spectrum is not the same as an FFT. A shock response spectrum can be thought of as the maximum absolute response of a series of single degree of freedom oscillators to a given base motion. Try doing a google search on shock response spectrum calculation. Another thing to keep in mind with SRS analysis is that while it is less computationally intensive than a transient analysis, the results of a shock spectrum analysis are statistical estimates of the structural response and not the actual response. This is due to the way shock spectrum analysis works. For linear systems, the complete response to a transient input can be described as the linear combination of responses of each mode of vibration. The individual modal responses are solved for and then combined to get the total response. All modes are assumed to act in phase, hence the predicted stresses may be higher than actual stresses. Direct linear combination and square root sum of the squares are two methods used to combine the modal responses, with direct combination being the most conservative. Also, if you are interested in anything other than the peak response, you can't get that from the SRS analysis either.
