Hi I am Arnold Greyling from Johannesburg in South Africa. This is my peer review assignment for week 2 of the Introduction To Music Production course at Coursera.org. I will be discussing the Analog to Digital Conversion Process.
Analog to Digital conversion is converting a physical value (voltage) to a digital value that represents that physical value.
The following aspects of the Analog to Digital Conversion Process will be covered:
The Analog to Digital Conversion Process is also known as sampling.
Representation of data in a computer
Data is always represented in a computer as a collection of binary bits. A bit represents a value of “0” or “1”. The bits are combined into bytes or words representing numeric values. A byte is normally understood as 8 bits. A word may have different lengths and the number of values that can be represented by a word is dependent on the number of bits in the word. The number of bits in the word is also known as word-length. The number of values is always the power of two of the word-length. See the table below for the number of values represented by often used word-lengths.
Choosing the correct word length
The word-length represents the number of discrete values that can be represented over the range of the analog value. There will always be an error in the conversion from analog to digital. The larger the word-length the smaller the error will be. A large word-length also allows a higher dynamic range in the converted value.
You always want to choose a high word-length to reduce the error and increase the dynamic range. A high dynamic range allows you to record at a slightly lower lever to prevent distortion due to clipping cause by a too high level.
The CD standard uses a 16 bit word-length which represents a dynamic range of about 90dB. You should choose at least 16 bits if you are planning to do high quality recordings. It is better to use 24 bits or if your Digital Audio Workstation allows it 32 bits. At 24 bits your dynamic range will be about 120db.
A single word in the computer represents a single analog value. Sound however changes over time and therefore we need to sample the changes over time. The rate at which we sample the analog value is known as the sample rate or sample frequency. To be able to fully recover a analog signal from a digital representation of it the sampling frequency has to be more than twice the maximum frequency of the analog signal.
The frequency range of human hearing is 20Hz to 20Khz, this means that the sampling frequency used to sample sound must be higher than 40 Khz.
Choosing your Sampling Frequency
Choosing your sampling frequency is quite important when you start a project because it may have consequences later on if you chose incorrectly.
The CD standard has a sampling rate of 44.1kHz so if you are recording for sound production only this will be a good choice. Professional video equipment normally sample sound at 48kHz. If you are recording a soundtrack you should choose 48kHZ to stay compatible with the video equipment. If you are in doubt choose the highest sampling rate available to you
The above gives a simple overview of the analog to digital conversion process. When you consider choosing your word-length and sampling frequency you should remember that higher values for both will increase the amount of data storage you need. Sometimes you may have to choose a lower word-length or sampling rate due to storage limitations.
If you want to go into more detail regarding analog to digital conversion a good start is at Wikipedia.