Hey guys. Sorry for long absence: running a recording studio in Doha proved an all-consuming challenge. Finally got the time to write another post! This time I want to write about sample rate and bit depth. But first...
Few weeks ago, I had a debate with some colleagues about the sample rate, bit rate, and which ones are best for a professional, competitive sound.
The opinions divided, and many of us started to defend what we thought was right. Too many perspectives. Everyone thinks they’re right, but we don’t research enough to find evidence to prove the point. In the end, we all learned a lesson…or two. We must be cautious when it comes to getting information related to sound recording and music production from the Internet. There are many people out there harming the new comers with misinformation, and even the guys with some years of experience giving out false information just because at both ends they lack the fundamental understanding of sound recording and sound production.
Since that day, I promised I would always ask my mentor to confirm all the new knowledge I acquire from a particular source. And a personal advice: study, guys! That's the best and fastest way to succeed.
So let's begin.
What is Sample Rate?
This is an excerpt from an article published on the Sonar works website. Quote: “The sample rate of an audio file describes the frequency response of the captured audio, and we want to reproduce frequencies up to at least 20 kHz. To reproduce 20 kHz, the sampling rate must be at least double the frequency, so a minimum sampling rate of 44.1 kHz must be used.”
In other words, sample rate means how many times per second a measurement is taken when converting an analog audio waveform to a digital signal. It defines the frequency response of an audio recording. Let’s look now at a Theorem called the Nyquist-Shannon Sampling Theorem, discovered simultaneously by Vladimir Kotelnikov, E.T. Whittaker, Harry Nyquist and Claude Shannon. It is also known as the cardinal theorem of interpolation. It says that the highest frequency we can record is half of the sampling rate. Therefore, when we record at a sample rate of 44.1 kHz, it actually means that we can only record audio signals up to 22.05 kHz. So, if we record a 96 kHz sample rate, we are getting 48 kHz of audio bandwidth. On the other hand, if we try to record above half the sample rate, or if we attempt to break the rules of the theorem, audible artifacts called aliases occur. But don’t panic! In the Analog to Digital converters there is a low pass filter called anti-aliasing filter which eliminates aliasing by using a low pass filter on the analog signal at half the sample rate.
Since we can hear up to 20 kHz, we can record at a sample rate of 44.1 kHz as it captures all the audible bandwidth for humans. However, the anti-aliasing filter can negatively affect audio below 20 kHz. Therefore, it is recommended to produce and mix at 48 kHz for popular music and 96 kHz for jazz, classical and world music, for example. Why 48 and 96 kHz? The first one, 48 kHz allows for better sounding anti-aliasing filters than 44.1, it uses a little bit more disk space than 44.1 and also because videos usually require 48 kHz audio. And the 96 kHz, its logically beyond this point. It eliminates audible high frequency aliasing and filter-induced distortions, it can provide lower processing latency and other features. In addition, a 96 kHz recording can be reduced to 48 kHz when needed. So thats the answer there!If your DAW gives you the option to record at 88.2 kHz, well…. That’s another good option to consider.
Stay tuned for the next post about the bit depth!