Adobe AUDITION 3 User Manual page 180

Provides the following options:
Gain Processor
• Specifies the amount of gain (measured in decibels) added to the output signal.
Output Gain
• Determines how long the processed output signal takes to reach the specified output volume. For
Attack Time
example, if a portion suddenly drops 30 dB, it takes time specified before the output drops to its corresponding
volume level. If the sum of the Attack and Release times is too short (less than 20 milliseconds total), audible effects,
such as a vibrating sound, occur at around 1000 Hz/milliseconds total. For example, if the Attack and Release times
are each set to 5 milliseconds (10 milliseconds total), then a vibrating sound occurs at 100 Hz. A total of 30 milli-
seconds is about as low as you can go without introducing these effects.
•
Determines how long the previous output level takes to reach the specified output volume. For
Release Time
example, where the Attack Time is how long the start of a pulse takes to reach the desired output volume, the Release
Time is how long the end of the pulse takes to reach the desired level.
•
Uses both channels to find a single input decibel value and amplifies both channels by the same
Joint Channels
amount, preserving the stereo center-channel image. With stereo files, each channel can be compressed indepen-
dently, sometimes causing the surrounding background noise to get louder on one channel. For example, a loud
drumbeat in the left channel makes the background noise in the right channel louder than in the left.
Provides the following options:
Level Detector
•
Specifies the amount of gain (measured in decibels) added to the signal before it goes into the Level
Input Gain
Detector (the section that detects the current level). The input gain essentially moves the graph plot up or down.
•
Determines how long the processed output signal takes to reach the specified output volume. For
Attack Time
example, if a portion suddenly drops 30 dB, it takes time specified before the output drops to its corresponding
volume level. If the sum of the Attack and Release times is too short (less than 20 milliseconds total), audible effects,
such as a vibrating sound, occur at around 1000 Hz/milliseconds total. For example, if the Attack and Release times
are each set to 5 milliseconds (10 milliseconds total), then a vibrating sound occurs at 100 Hz. A total of 30 milli-
seconds is about as low as you can go without introducing these effects.
• Determines how long the previous output level takes to reach the specified output volume. For
Release Time
example, where the Attack Time is how long the start of a pulse takes to reach the desired output volume, the Release
Time is how long the end of the pulse takes to reach the desired level.
•
Is a mode provided for backward compatibility. It is a graph interpretation method that is slightly outdated
Peak
and a bit more difficult to use than RMS. It is equivalent to twice the RMS value (for example, -20 dB in RMS mode
equals -40 dB in Peak mode).
•
Is a graph interpretation method that more closely matches the way people hear volume. This mode causes
RMS
the output to be exactly the RMS amplitude that is specified in the graph. For example, a limiter (flat horizontal line)
at -10 dB causes the RMS amplitude of the result to average -10 dB (where 0 dB is a maximum amplitude sine wave
without clipping).
Specifies the lowest frequency that dynamics processing affects.
Low Cutoff
Specifies the highest frequency that dynamics processing affects.
High Cutoff
Helps to handle sharp spikes that might occur at the onset of a louder signal by starting the attack
Lookahead Time
the specified number of milliseconds before the audio becomes loud, instead of right on top of the transient. The
spikes occur because of the time needed to determine and react to the current signal level (as determined by the
attack values). For brief moments, these transients can go beyond the limits of the compressor settings. You might
want these spikes in certain compression scenarios to enhance the impact of, say, a drum hit, but they aren't desirable
if you're using limiting to reduce the maximum amplitude. Otherwise, with Lookahead Time set to 0, a spike stays
loud until all of the attack times elapse.
ADOBE AUDITION 3.0 175
User Guide