WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions.
Click here to have your audio equipment measured for free!
I made the downsampling and frequency spectrum changed , and i mean in the audible range .Especially 20-100 hz looks much smoother in hires file. The waveform seems the same though. Why is that ?
The frequency resolution of FFT depends on the total duration of the samples (Δf = 1/T = fs / N). When you halve the sampling rate but stay with the same number of samples, your frequency resolution is doubled. That's why the peaks and dips at the lower frequencies looked less smoothed with the lower sampling rate.
Same FFT size for more than twice the samplerate. So less detail per freq bin (broader range of freq sampled in 1). Also same graph width where one runs till 22 and the other till 48kHz. Scaling is therefore different and graphs cannot be overlaid or easily compared side to side.