REVIEW ZONE - Audio Compression : the new generation ( Page 2/12 )

Home

Hardware

Audio
Controllers
General
Mainboards
Modems
Monitors
Portables
Printers
Processors
Scanners
Storage
Video

Games

Reviews
Previews
Cheats & Hints

News

Current News
Archives

General Info

About Us
Advertising
Awards
Contact Us
Hot Links

Downloads

Message Board

The basics of audio compression

If you've ever tried to ZIP a .WAV file (or another generic file compression scheme) you would have noticed that the file size hardly shrinks. Then how do audio compression techniques manage to squeeze these hard-to-compress files by as much as a 10:1 (or more) ratio? And why can't we use those same techniques to really shrink our other documents and programs? The answer is the concept known as "lossy" compression. When we uncompress a program file or a document we want the resulting file to be identical to the original. A minuscule error can easily result in a non- working or worse, an erroneously working program. When it comes to a high quality audio recording this is not so. There are many sounds that are beyond the range of human hearing. Compression is achieved by removing this extra information. For example, the human ear loses its sensitivity to sounds which have a frequency higher than 16 kHz, at 20 kHz dropping to almost no sensitivity. It is common practice to consider 22 kHz the upper limit due to technological limitations.

It is not only the higher frequency sounds which can be removed from audio to reduce storage space. The human ear also has certain "psycho-acoustic masking" effects. This is just a fancy way of saying that certain sounds block other sounds. For example,
a very loud note can block other notes whose frequencies fit certain conditions. This masking effect also occurs over time-
after certain sounds, other sounds will be blocked out for a
period of time. This is, of course, a gross simplification and
there is a lot of research going into this field.

When trying to compress stereophonic music to small sizes
(low bandwidth) another compression technique uses the fact
that there is a lot of redundancy in the left and right channels.
By encoding the audio common to both channels together and
the differences separately, stereo compression achieves better efficiency than mono compression.

Frequency response
Though the human ear can only recognize frequencies up to 20 kHz, it is often thought that people can still "feel" higher frequencies and this is considered lacking even in CD music. The upper frequency limit of the ear also drops with age and varies from person to person.

Bandwidth and file size
Bandwidth is a often used to describe the compression level of audio. 128 kbps, for example, means that the compressed audio takes 128 thousand bits for one second of audio.

If you wish to learn more about the way audio compression works see the links section following the conclusion (last page).

The way audio is compressed invalidates some of the normal techniques used to compare audio equipment. Let's look into that next.

Table of contents

Introduction TwinVQ (VQF)

Basics of audio compression Real Audio G2

Methods of comparison Microsoft Audio 4

Software and hardware MPEG 1 layer 2 (MP2)

Selection of music for the comparison MPEG 1 layer 3 (MP3)

Index of results Conclusion