Although I learnt that sample rate means how many samples do you get per second, does it really makes a lot of difference when recording music? Most of the instrument we use is no more than 2000hz, is 96khz too much according to Nyquist theorem? Let's do an experiment! Instrument: Guitar Recording:Zoom H6 DAW: Studio one The signal showed here is about the same time length. Experiment 1 Playing 6th guitar string openly, in key E2, base frequency around 70hz: Samplerate: 96 kHz pretty smooth, only at steep part can see a little bit jag. Samplerate: 44.1 kHz You can still tell the shape of the curve, not very smooth, you can see jag almost everywhere. Experiment 2 Playing Overton at 1st guitar string at 12th fret, in key E6, base frequency around 1200hz: Samplerate: 96 kHz On the steep curve, you can see jag and noticeable jag everywhere. But the signal still seems smooth enough and you can clearly tell the shape of the curve. Samplerate: 44.1 kHz Jag everywhere, so obvious. Can hardly tell the shape of the original signal. So, for high-frequency sounds seems we still need to use something like 96khz or even higher sample rate to achieve a better signal quality. Experiment 3 But what if I mix down the 96kHz file to a 44.1kHz file? Will it still be same worse as using 44.1kHz directly? This graph shows the signal of the 1st picture in Experiment 2 which converted to 44.1kHz. you can see a lot more jag than the 1st picture of…