hello, everyone, it's 
Sampling:
This the act of capturing an amplitude value over time. A sample is a static representation of a waveform in time. The number of samples taken per second is called the samplerate.
for instance; if we sample a 20 kHz frequency using a sample-rate of 20kHz, we only take one sample per completed cycle. To capture both the positive(+) and the negative(-) side of the waveform we will need to use a sampling rate that is double that of the frequency we want to sample.
In other words:
To sample a 20 kHz frequency, we will need to use a samplerate of at least 40 kHz.
N/B: That this is the basics of the Nyquist theory by Henry Nyquist which is written as:
FS = 2 X FN
Where:
FS = Sampling frequency
FN = Nyquist Frequency
Anti-aliasing filter.
The highest frequencies that we try to sample will create some aliasing errors and other problems and this can be properly captured. only to filter off the highest frequencies that we are sampling with a low-pass filter and this is the reason we will need to get a higher sampling rate than 40 kHz to properly capture a 20 kHz frequency. Keeping in mind that the cut-off frequency of the filter is around 20 kHz so we don't lose too much of the original signal.
So we can say that sampling is capturing an amplitude value in time. This value is captured in a process called quantization
Bit-depth:
As well as the time axis, digital systems have a resolution on the amplitude axis as well. The higher this resolution, the more accurate the digital representation of the waveform amplitude. This is determined by the bit-depth of the recording.
The higher the bit-depth, the bigger the dynamic range is that can be captured.
So a 24-bit system has a higher resolution than a 16-bit system.
N/B: A 16-bit recording has a theoretical dynamic range of 96 dB while a 24-bit recording has a theoretical dynamic range of 144 dB.
You can imagine that mixing with audio that has a 24 bits resolution is much easier and more pleasant to mix with when compared to 16 bit since we literally have more values on the amplitude axis
So why a 44.1 kHz sampling rate?
This sample rate standard dates back to the days of video-cassettes. It was chosen because it was higher than 40 kilohertz(needed to sample a 20-kilohertz signal). and it was mathematically compatible with both the European PAL, as the US-NTSC video format standards.
This 44.1 sample-rate with 16-bit bit-depth was also adopted as the audio quality for the standard audio CD'S. This sample-rate has been relevant for years and up to today it actually still is...
common sample-rates that are being used nowadays during mixing can be a bit higher.
Not only does this means that we can theoretically record higher frequencies, but it also means that we have a lot more samples that are taken to represent our audio.
N/B: Many operating systems and audio interfaces nowadays can support sampling rates up to 96 kHz or even up to 192 kHz. And this is giving a much higher audio quality and processing options.
The filter and the AD-converters that are being used by an audio interface are really crucial in determining what the quality of the interface is that is actually doing these conversations.
The sample-rate that we are using for our project can be seen as it's playback speed. Audio files in a project at are on the currently running sampling-rate will playback at a normal speed, but if we force to playback, a 48 kHz audio file in a project that is running at 44.1 kHz, we will hear a slowed down version. played back slower, in a lower pitch. and if we playback audio files in a lower sample-rate than of the recording, it will play back faster, and pitched up.
N/B: Recording on higher sample-rates and bit-depths will create higher resolution audio, and thus produces larger file sizes.
We can quite simply calculate the amount of disk space needed for recording our audio at various sample-rates and bit-depths. and this will be done as
- Sample-rate x bit-depth = amount of bits per second
- amount of bit per second x duration of recording = bits for duration of recording
So let us calculate this for a song of duration of 4 minutes(240 seconds), 44100 hertz, 16 bit - 44100 hertz x 16 bit = 705.600 bits per second.
- 705.600 x 240 = 169.344.000 bits per 240 seconds.
We can easily convert the amount of bit per seconds to megabyte per second by dividing the total value of the bits per seconds by 8 - 169.344.000 bits / 8 = 211.68.000 bytes per 240 seconds.
Also diving this value in bytes per second by 1024 will give us the value in kilobyte per second. - 21.168.000 bytes / 1024 = 20.671,875 kilobytes per 240 seconds.
diving this value in kilobyte per second by 1024 again and we will have the value in megabytes per seconds. - 20.671,875 kilobytes / 1024 = 20.1 megabytes per seconds. this is the result of 4 minutes of digital audio at 44.100-hertz sample rate and 16 bit multiply that by two(2) and you will get the number of megabytes needed for a stereo track.
Comparing the filesize of 1 minute of audio for various samplerate and bit-depth settings.
- 1 minute - 16 bit - 44100 Hz - 5.04 Mb
- 1 minute - 24 bit - 48000 Hz - 8.28 Mb
- 1 minute - 24 bit - 92000 Hz - 15.79 Mb
- 1 minute - 24 bit - 196000 Hz - 33.64 Mb
You can see that the higher the samplerate and bit-depth are the larger the files size get.
Reasons for choosing higher samplerate
They are a number of reasons why we would want to choose to work with higher samplerates and bit-depths and this include:
- The better quality that higher samplerate gives and the more dynamic range than the higher bit-depths gives. Especially the higher bit-depth which is really pleasing to work with when mixing.
- Processing is also a lot more accurate on higher resolution material. EQ'S, REVERS, FILTERS, etc. they all sound much better on a higher samplerate material. This is the reason why many plug-ins use the so-called up-sampling techniques to internally work at higher samplerates when processing audio.
- Another reason to work on higher samplerate is when you need to apply time stretching or maybe even pitch correction. Because we have so much more samples on higher sample rates, we have a lot more accurate information about the frequencies and that means that we can do a lot more accurate stretching and correction.
- And another reason to work at higher samplerate is for example when we are producing sounds for DVD'S. DVD'S can handle samplerates up to 192 kHz on stereo materials and 96 kHz for surround materials.
Common samplerates
44.100 Hertz - Audio Cd quality
48.000 Hertz - Pro video, Dat audio, sampling rate of a lot of professional audio gear.
88.200 Hertz - not often used doubling of the Cd-quality, can sometimes be found.
96.000 Hertz - DVD audio quality supporting stereo mixes up 5.1 surround mixes.
192.000 Hertz - DVD audio quality supporting stereo mixes.
N/B: Working on higher bit-depth doesn't stretch your computer out too much. however, working on higher samplerate will stretch your computer a lot more. So you will be needing much more processing power if you don't have that.
Also, keep in mind that you will not be able to add as many tracks and as many plug-ins as you were used to when you work at lower samplerate.
And now to answer a much-asked question, If we're going to create a mix that is going to end up on a CD, aren't we going to be mixing it down to 44.1 at 16 bits eventually??
Well, that answer is , YES, YOU ARE.. but that only in the very last stage after the mastering. So during mixing and during recording, we can work at this higher samplerates and bit-depth, and only in the very last exporting stage of the master, we are going to sample it down to 44.1 and 16 bit if you are going to be burning this to a CD.
That will be all for now, If you missed our previous editions you can check it out through these links: REF 1 REF 2 REF 3 REF 4
And like always you are free to send in your questions and suggestions on the comments section below, and I promise to do my best in replying to them.
Thanks for reading!!!!
Until next time!!!