featured artists musiq.com inteligence in music
featured artists the recording studio machines and meaning modes and melodies ancient modern lutes wise shopping  
Recording: digital audio Digital Audio Tutorial: Introduction to digital audio

Digital audio tutorial
Introduction to digital audio
What sounds best?
Music compression
Common computer digital audio formats
Perceptual encoding schemes
Chart 1: bit depth and sampling rate chart
Chart 2: audio file formats
Chart 3: compression schemes
Glossary of common digital audio terms
 
Jump to:
mp3 tutorial
Manufacturing links:
Outboard Gear
Mixing Consoles
Microphones
Recording Devices
Amps and Monitors
musiq.com digital audio
Digital audio is a stream of bytes which contain amplitude (volume) data. In other words, when you're talking about samples, or a CD, you're talking strictly about a series of bytes which represent a sequence of volume peaks. How is it, then, that in digital audio, one can hear frequencies?

Frequency, or what we hear as "pitch" or "timbre," is a result of periodic repetition of an amplitude. If an amplitude is represented by a sample 220 times in a second, we hear a 220 hz (low A) frequency. Consequently, a second of digital audio can represent any combination of frequencies, to the limit of both the sampling rate (how many of these amplitudes per second) and the bit depth (how much data represents each amplitude).

Taking standard CD audio, we have a format which transmits 44,100 samples per second with a bit depth of 16 bits. Starting with the bit depth, a 16 bit byte can represent numbers from 0 to 2^16 (65536). In comparison, an 8 bit byte can only represent numbers from 0 to 2^8, (256). It should be obvious the huge difference made by CD audio (at least over old 8 bit sampling keyboards).

With a sampling rate of 44,100 hz, you might wonder why CD audio is only capable of frequencies up to 20,000 hz. There is a theorem, called the Nyquist theorem, which states that the highest audible frequency is half the sampling rate. This is because sound waves are perceived due to the alternation of peaks and troughs. Without hearing the full wave (peak-trough-peak-trough), we don't perceive the existence of that frequency.

Technically, then, a CD should represent frequencies up to 22,050 hz. Due to "technical difficulties," raw digital audio contains a huge amount of distortion in the top 3,000 hz., due to the frequencies higher than 22,050 which are partially represented, and due to mechanical flaws which produce "jitter." As such, audio engineers developed oversampling, which is a filtration system for high-frequency jitter and distortion, effectively obliterating the top 2000 hz. Hence, 20-20000.

This isn't the whole picture, however. We're supposed to believe that since human ears pick up sound primarily in the 20-20000 hz range that CD audio is thus the ultimate medium for sound. There's a new standard on the block, called DVD-audio, which is 24-bit 96khz. digital audio. 24 bit means 16.7 million amplitudes, and frequencies up to 48 khz. And it definitely sounds better!

    read more on CD vs. DVD...
 

musiq.com news links about musiq.com contact musiq.com © 1997-2004 miq productions