I have to correct myself. The function test_fft1() still does not work, it just looked good in the plots, but at a closer look, Re(IFFT) is close to zero and is far from matching the exact IFT.
So it seems FFT(IFFT(f)) == IFFT(FFT(f)) == 1 (if done right ;-), but I just
cannot reproduce the exact (I)FT.
Felix
_______________________________________________
Numpy-discussion mailing list
[email protected]
http://projects.scipy.org/mailman/listinfo/numpy-discussion
