|
E-mail us with your comments and feedback about this article. |
This document provides detailed information about how dynamic range control, a new feature in Microsoft Windows Media 9 Series, is implemented.
|
|
Jared Davis Microsoft Corporation November 2004
Applies to: Microsoft® Windows Media® Encoder 9 Series
IntroductionThe Microsoft® Windows Media® Audio 9 codecs include a new set of features for content providers, including high resolution audio, multichannel capabilities, author-controlled fold-down coefficients, and dynamic range control. The purpose of this article is to explain the use and implementation of the dynamic range control feature in the Windows Media Audio 9 Professional and Windows Media Audio 9 Lossless codecs.
Dynamic range is defined as the difference, in decibels (dB), between the loudest and quietest sounds in any particular piece of audio content. Classical music is a good example, with ranges from piano (soft) to forte to FFF (for extremely loud). Movies also typically have a wide dynamic range, which may cause you to have to turn the volume up and down as scenes change. For example, when watching a movie at home, you may be forced to turn up volume to hear the dialog in a quiet scene, and then quickly turn it down again during a car chase scene that follows. In this way, there may be times in a home theater environment when it would be useful to be able to control the dynamic range.
 Back to Top
Dynamic Range Control in Windows Media 9 SeriesWhen content is encoded using the Windows Media Audio 9 Professional or Windows Media Audio 9 Lossless codec, the codec calculates dynamic range values and places them in the header of the encoded Windows Media file. The values calculated during encoding are the average loudness and the peak audio level of the content. These values are referred to as the reference average and reference peak values, respectively.
During playback, dynamic range control is implemented by the Quiet Mode feature in Windows Media Player 9 Series and later. The Quiet Mode feature has three settings that affect dynamic range: Off, Medium difference, and Little difference. By default, the settings affect the dynamic range during playback as follows:Off. Content is played with the full dynamic range without any processing. Medium difference. Content is played with a medium difference between loud and soft sounds, meaning less content compression is used. The peak value of the audio signal is limited to 12 dB above the average level. Little difference. Content is played with very little difference between the loud and soft sounds, meaning the most content compression is used. The peak value of the audio signal is limited to 6 dB above the average level.
Back to Top
Adjusting Dynamic Range ValuesWindows Media File Editor, a utility included with Windows Media Encoder 9 Series, enables content creators to specify different dynamic range values than the reference values that were calculated during encoding. When you open an encoded file in Windows Media File Editor, you can view the reference peak and reference average values and specify different values for each. For the purpose of this article, values that you specify by using Windows Media File Editor are referred to as the target peak and target average values. Note that specifying target peak and target average values does not overwrite the reference values; instead, the target values are added to the header of the Windows Media file.
This section provides an overview of best practices and expected behavior when specifying target peak and target average values by using Windows Media File Editor. It is typically recommended that you only adjust the peak value. Adjusting the average value does not affect the difference between loud and soft sounds. Instead, it cuts or boosts the overall average volume of the entire piece of audio, which may cause undesirable distortion during playback.
If you specify a target peak value (but do not specify a target average value) by using Windows Media File Editor, the Quiet Mode settings affect the playback as follows:Off. Content is played with the full dynamic range without any processing. Medium difference. The peak audio signal of the content is limited to the target peak value, unless the reference peak value is less than the target peak value. If this is the case, the peak audio signal is unaltered. The average volume matches the reference average value. Little difference. The peak audio signal of the content is limited to the mean of the target peak value and the reference average value, if the mean is less than the reference peak value. If it is not, then the peak audio signal is unaltered. The average volume matches the reference average value.
For example, assume the reference peak value is -3 dB and the reference average value is -20 dB. If no target peak value is specified, the Medium difference setting in the Quiet Mode feature yields a peak value of -8 dB, while the Little difference setting yields a peak value of -14 dB. However, if you specify a target peak value of 0 dB by using Windows Media File Editor, the Medium difference setting yields a peak value of -3 dB (the reference peak value), and the Little difference setting calculates the mean between the target peak and target average values to yield a peak value of -10 dB. (The mean is calculated by adding 0 dB and -20 dB, then dividing by 2.) On the other hand, if you specify a target peak value of -10 dB, the peak for the Medium difference setting is -10 dB (the target peak value), and the peak for the Little difference setting is -15 dB (-10 dB plus -20 dB, divided by 2).
If you specify both a target peak and target average value by using Windows Media File Editor, the playback behavior differs depending on how the original dynamic range (reference peak minus reference average) compares to the target dynamic range (target peak minus target average). Keep in mind that specifying a target average value is not recommended. If the original dynamic range is less than the target dynamic range, then the original dynamic range is used. However, if the original dynamic range is greater than the target dynamic range, then the Quiet Mode settings affect the playback as follows:Off. Content is played with the full dynamic range without any processing. Medium difference. Content is played back using the target average value; the peak value is limited to the target peak value. Little difference. Content is played back using the target average value; the peak value is limited to the mean between the target peak and target average values.
For example, assume reference peak and average values of -3 dB and -20 dB, respectively, and target peak and average values of -5 dB and -15 dB, respectively. Since the original dynamic range (-3 dB minus -20 dB equals 17 dB) is greater than the target dynamic range (-5 dB minus -15 dB equals 10 dB), the Medium difference setting yields an average of -15 dB and a peak of -5 dB. The Little difference setting yields an average of -15 dB and a peak of -10 dB.
As another example, assume reference peak and average values of -3 dB and -20 dB, respectively, and target peak and average values of -5 dB and -25 dB, respectively. Since the original dynamic range (-3 dB minus -20 dB equals 17 dB) is less than the target dynamic range (-5 dB minus -25 dB equals 20 dB), the Medium difference setting yields an average of -25 dB (the target average), and a peak value of -8 dB (-25 dB plus 17 dB). In the Little difference setting, the target dynamic range is equal to half the target dynamic range of the Medium difference setting. Therefore, in this example, the target dynamic range of the Little difference setting is 10 dB (half of 20 dB), so the average used remains -25dB, but the peak used is -15 dB (-25 dB plus 10 dB).
Back to Top
For More InformationTo learn more about implementing dynamic range control in Windows Media 9 Series, see Windows Media File Editor Help. Windows Media File Editor is included with Windows Media Encoder, which can be downloaded from the Windows Media page at the Microsoft Web site.
Back to Top
|
