Dynamic Processing in Music Production

Dynamic Processing

• Dynamic processing in music production refers to technical microdynamics, not tone strength.
• It involves four control directions: compression, limiting, downward expansion, and upward expansion.
• Compression (A) increases the ratio above the threshold.
• Limiting (B) has a very high ratio above the threshold.
• Downward Expansion (C) reduces the ratio below the threshold.
• Gate (D) acts as a switch, controlling below the threshold.
• Upward Compression (E) controls below the threshold.
• Upward Expansion (F) controls above the threshold.
• Processing occurs above or below a threshold.
• Dynamics may be increased or reduced.

Dynamic Behaviour (Attack Time)

• Dynamic processors aren't instantaneous, requiring time to adjust to input levels (attack time).
• Attack time is the delay between exceeding the threshold and reaching 63% of the target value.
• Typical attack times range from 0.1 to 50 milliseconds.
• Attack time is not a waiting period, but rather a settling time.

Release Time

• Release time is the time it takes for the processing to return to its initial state after the input falls below the threshold.
• Release time is typically slower than attack time, ranging from 10 milliseconds to 3 seconds.
• Release time is crucial for sound character in various devices.
• Extreme attack and release settings may negatively impact sound quality, potentially interfering with signal oscillations.

Static Behaviour

• Static behavior describes the steady-state response after the attack time.
• Control amplifier level changes can be visualized using characteristic diagrams.

Soft Knee

• Soft knee compression has a gradual onset, smoothing the transition to compression upon reaching the threshold.
• Normal or hard knee compression rapidly activates compression at the threshold.
• High compression ratios can lead to audible compression around the threshold, a less desirable effect.

Control Type & Sidechain

• Control amplifiers are similar across applications.
• Input signals use rectifiers and detectors to compare to a setpoint, generating a control voltage signal.
• Feedback and feed forward control pathways are common.

Sidechain

• Sidechain input processed signal can be used to control the main signal.
• Filters within the sidechain may be fixed for certain tasks.
• This affects the controlled variable in the main signal.
• Example: adjusting the bass response.

Look Ahead

• Look Ahead delay allows the detector to react to changes faster than they occur.
• Ensures consistent handling of signal peaks—especially in Brickwall limiters.
• Improves peak capture reliability.

Gate and Hysteresis

• Gates and expanders are often equipped with hysteresis.
• Hysteresis is another threshold based on the first threshold.
• This provides a safety margin, preventing premature processing of rising signal levels.
• With hysteresis, the signal needs to drop below a second level before processing occurs.

Peak vs. RMS

• Detectors can respond to either signal peaks or RMS values or energy levels and this impacts the output response.
• Crest factor describes the signal's peak-to-average level.
• RMS detection tends to be more constant.
• Peak to RMS-based processing strategies often are employed based on the need to smooth or reduce the apparent effect of peaks in a signal.

Parallel Compression

• Parallel compression processes a signal component in the audio path and mixes it into the audio's original.
• Signal is split and one copy processed, mixing and returning the results to the signal output.
• Processing should be within very tight tolerances, in terms of timing between processed and original signals.
• This method can be used for adjusting the level, especially in situations with widely differing peaks and average signal levels.

Multiband Compression

• Multiband compression divides the audio signal into frequency bands, allowing independent processing across each band.
• This allows different levels of compression for different frequency bands, producing a more nuanced sound.
• Multiband compression usually contains a series of crossover filter outputs generating separate input signals to independent signal processing stages, and then recombining these into an output signal.