There's a big difference between just showing a bucketed Fourier transform of the whole audible spectrum, and showing a high-resolution, log-scale, musically-informed visualization of the part that can be heard as musical notes.