Please note - you can only use Fixed Point representation with the Variable Streaming FFT.
The FFT essentially doesn't care about your fractional part - it assumes the input is an integer, and you'll get an integer result (with some bit growth as described in the user guide), of course - due to the way these things work, you can use it with an implied fractional part. So for example, if you have an 18 bit input and that's got 10 fractional bits you'll get an output width 29 bits (with bit growth at the top end) - 10 of which are fractional.
