Part 2: How Computers Store/Play Sound3
Computers store sound digitally, in 1s and 0s, rather than the actual sound. To show how computers store sound, we will use the graphical representation of sound.
A sound file has 2 primary properties: its frequency (in Hz), and its resolution (in bits). The frequency is not related to the frequency of a note, but refers to the number of times a computer will check the sound level. The most common frequencies are 22,050 Hz (1 Hz=1 Time/Second), 32,000, and 44,100 (CD Sampling Rate). At each check, the sound inputs a volume in the form of a whole number (1,2,3,...). Because computers store numbers in binary form (1's and 0's), the bit number refers to the number of digits used. For example, a 2-bit resolution would have four levels: 00 (0), 01 (1), 10 (2), and 11 (3). A 4-bit resolution has 16 levels, etc.
Example of Sampling Rate/ Converting to Binary (They use Voltage instead of volume-the two are proportional)4
The number of levels for an X-bit resolution is equal to 2X, so a higher bitrate equals a higher quality of sound. The most standard form for sound is 16-bit (65,536 levels), with older files having 8-bit resolution (256 levels). Frequency and bitrate determine the size of the file-each frequency mark contains the number of bits (8 bits is a byte). 3 seconds of a 44,100 Hz, 16-bit file is 16(bits per reading)x44100(readings per second)x3(seconds)=2,116,800 bits (/8 bits per byte)=264,600 bytes (264 kilobytes).
To play back the sound, the computer plays a tone at the volume it has from the data file. because theses tones go so fast (44,100 of them a second), we cannot hear them-we only hear the sound that was recorded into the computer. To represent this graphically, we could plot the volume along a y-axis and the time on the x-axis. By making straight lines between the dots, we see the basic "wave" the computer plays.
Recording Wave, Sample Taken by Computer and Resulting Wave4
Consequences of Digital Recording/Playback
While digital music allows sound to be manipulated and copied very quickly (as opposed to analog recordings, where a track had to be played in its entirety in real time to be copied or transferred with an effect applied), the means of converting to digital open the sound up to quality problems. For example, if the sampling rate is the same as the frequency of a wave, all that would be sampled is a constant level, which would produce no sound. While most sound files are sampled above the range of human hearing (28 kHz, with the max human range being about 20 kHz), and most problems are more complex due to complex sound waves, the problems loss of resolution can cause are still important, and appear more with compressed sound.
Part 3 - Sound Compression and Formats