Dynamics processing is simply what the name implies ... manipulating the dynamics of an audio signal. The two processes with which we are most familiar with are compression and limiting. Compression and limiting involve automatically lowering the signal as the level of one signal increases, thereby reducing, restricting or limiting the dynamic range.
On the surface, a dynamic controllers appears to be a fairly simple device. All the action takes place in the gain control element with an amplifier on the front and back end to match up signal levels to the outside world. There are several different devices that are used as a variable gain element. To a large extent, it is the characteristics of the different devices that give each compressor its "personality". Today, most dynamic controllers use a modular IC voltage controlled amplifier or attenuator (VCA) for gain control.
There are two paths in a dynamic controller, the main audio path and the side chain. The audio path is what you put in and what you expect to get out. But in order to derive a voltage used to adjust gain, the input signal must be detected and applied to the gain control input of whatever element is used for gain reduction. This path is called the side chain.
The method in which the side chain control voltage is derived is another contributor to the processors personality. Manufacturers have used simple averaging detectors, peak level detectors, and true RMS averaging detectors. Each method gives control voltage that follows the input signal a little differently. So, each type of side chain detector will impart a different control action on the main audio signal. In addition, designers have applied their own "corrections" to the side chain signal to compensate for non-linearity of the gain control element.
It should be clarified that audio dynamics processors can not operate instantaneously, as do conventional amplifiers. If they did, the processor would become non-linear which actually distorts the signal waveform. Obviously, we do not want waveform modulation in any audio system. What is desired is to alter the envelope, rather than the individual cycles themselves. This envelope is what makes the character of every processor on the market. How a processor deals with threshold, attack and release is what we call the envelope. Many people just see dynamic processors as a way to control levels of audio. When in reality, they are a far more creative tools in shaping the tone and impact of sounds. I will get into that later as well as some philosophy of sounds.
So how do processors deal with envelope? Ideally, for minimum signal distortion, the gain change would occur very slowly, following the general "shape" of the signal envelope, but not the actual cycle. Unfortunately, this can not be done in practice because of the need to control and deal with signals that have quick bursts of energy. When controlling signals, we require that the gain controlling mechanism respond relatively rapid to sudden applied bursts of an input signal, such as the attack from a drum or percussion instrument.
Once a burst of input signal has caused the processor gain to change quickly, we must prevent further instantaneous response to prevent the gain from changing with every peak and dip of the waveform itself. This "tracking" would effectively cause waveform distortion. A release type structure is needed in order to control how the gain returns to nominal. The time given to a release parameter controls how close the processor follows the envelope of the actual signal. Too fast of a release time can interfere with the actual signal waveform and cause distortion or bad modulation effects.
Now we come to the question: At what point in the signal level spectrum do these changes occur? This point is called THRESHOLD. Let us take the example of an infinity:1 ratio limiter. Let us say that a +4dBv input signal threshold is selected, and the processor has a nominal gain of 0db or unity. With such parameters, the processor will act as a normal unity gain amplifier as long as the input signal is lower than the specified +4dBv threshold. Since the processor is doing nothing under these conditions, its ratio is 1:1. Now let us suppose that an input burst measuring +10dBv. The limiter will "attack" in response to this over threshold level reducing its gain -6dB, causing the signal burst to exit the processor at +4dBv. The definition of threshold when applied to a limiter or compressor is a signal level below which the processor does nothing. Above a set threshold, a processor performs gain reduction according to its ratio.
The degree to which signals are altered in response to the threshold is called RATIO. Ratio, be it limiting ratio or compression ratio is expressed as the ratio between a signal level change at the input of a processor verses the signal level change at the processor output.. In a linear amplifier, the relationship of input change to output is direct. Thus, the ratio is stated as 1:1. A 1db increase at the input signal level produces a 1db signal boost on the output level. A "perfect" limiter has a ratio of infinity:1. So during limiting, an infinite increase at the input level is required to produce 1db at the output level. Therefore, the output is maintained at a constant "clamped" output level, and a "leveled" output results whenever the processor is limiting.
So how does dynamic processors detect these signals? Well, there are several alternatives available to circuit designers in structuring a detector mechanism. Peak detectors measure the voltage excursions or peaks of the polarity. Peak detection was the backbone of early processors which were designed for broadcast and disc cutting. Unfortunately, the electrical peak value of a music waveform has little to do with how loud the sound is perceived by the human ear due to the varying waveform complexities. If you were to take two musical waveforms of say a flute and a brass instrument, both would appear to have the same level of audible loudness, but exhibit completely different peak values. If the two waveforms were detected by a conventional fast peak detector, the brass waveform would be read as having a much higher energy level than the flute. If the detector were controlling a limiter, the audible result would be that the complex waveforms get over limited with respect to the simpler waveform. This is exactly what happens in many conventional dynamic processors. This is the effect most responsible for the squashed unnatural sound often associated with dynamic controllers.
In looking at most modern uses for dynamic control, it is found that the intent is not so much to control electrical peaks, but to manipulate the dynamics of the signal on the basis of the audio content.
Now days, equipment following the processor does not have a critical overload point, but instead headroom to accept the range of complex waveforms generally found in music and speech. Today, must studios operate at +4dBv as a nominal level but can handle signals much higher , getting into the +24dBv range. This headroom helps in handling complex waveforms other electrical peaks. Numerous psycho-acoustic tests have shown that when a exceedingly complex waveform that requires still more headroom appear, the inherent clipping of the extreme electrical peaks is of little audible consequence and is preferred over allowing these peaks to cause gain reduction in limiter type processors. Therefore, dynamic processors that are designed to perform audible-level dynamic control must employ a detection scheme which measures the audible content of the program to be effective. While true RMS detection appears to be ideal, there are several other factors which influence the human hearing, like harmonics. Harmonics of complex waveforms can be discriminated against with true RMS detection.
Another type of detector is the "classic" style photocell (light dependent resistor or LDR) as a variable resistor in a voltage divider. This type of detector uses a incandescent light bulb, LED, or electro luminescent (EL) panel. The photocells are driven by the voltage that comes from the input signals side chain (not the input signal itself). The signal goes through the LDR. The light gets brighter as the signal gets louder, causing the photocell to change resistance and in essence, turn a pot to reduce the gain. A Vactrol (R) is a sealed module consisting of a light source and LDR that was used in several "classic" compressors, and today you'll sometimes see the term "Vactrol" associated with a certain flavor of compression.
There are two major gain elements in use today. One type of gain control element is a vacuum tube that works as a variable resistor, lending the name "Variable-mu". U is the abbreviation for the tube's gain. The Fairchild 670 is one of the classic variable-mu compressor now selling on the vintage market for over $10,000. Altec also made one. Today, Manley Labs builds one using the same principle of gain control but with different tubes. They also make use of a full differential signal path which cancels out second harmonic distortion caused by the tube's action.
VCA compressors are most common today due to the availability of inexpensive and reasonably good VCA's in IC form. Since this style of VCA is all on one chip, the designer's challenge is to keep the signal that's controlling the gain out of the actual audio signal path. New designs have been pretty successful. The VCA is the fastest responding and most linear gain control element. It also lends itself well to the new breed of digitally controlled analog compressors. Here, the side chain control voltage can be shaped digitally to produce any imaginable response curve, allowing the compressor's gain control element to emulate the sound of any "classic" which can be measured. The Empirical Labs EL8 is one such compressor.
Just about everything inside the box affects the sound of a compressor - the gain control element, the way the side chain signal is derived and processed, the sound of the amplifiers in the input and output, and even the power supply. Today's much lusted-after "tube compressor sound" is really a new development. Of the much emulated classic compressors, only the Teletronix LA-2 had tube amplifiers. All the newer models in the series were solid state although they used the same basic (with seasonal variations) gain control elements. Much of what we think of as the warm tube sound of a compressor is a result of a tube input and/or output stage, not the compressing action itself. Certain compressors have a reputation for creating irregularities in frequency response, both favorable and unfavorable, but this in general isn't a result of gain reduction, but rather, input and output stages.
So why all the talk about envelopes and processor detection? The envelope holds a lot of the personality of a dynamic processor and can be used to give impact and punch. Depending on the envelope and the attack of a dynamic controller can make a instrument sound spanked or down right smashed. Other dynamic controllers really look into the harmonic content of the source material and can actually suck tone out and bring it forward without pumping.
So how can dynamic controllers help make mixes clearer and easier to listen to? I think its always important to consider what you are trying to accomplish. Are you just trying to control level or make something stand out or cut through a mix. Often times dynamic controllers can be used to give a mix support and power by keeping the low frequency content stable and solid. Or keep the mid frequencies from mowing over the high frequency detail of a mix. The more levels jump around the harder it is for the ear to understand and hear them. The more stable the level, the easier the ear can understand the tone giving things more clarity. For example the drummer that does a lot of snare work and has ghost notes playing through a performance. Dynamic controllers can be used to suck those ghost notes out and put them right where you want them in a mix. Other times it might be a vocal that needs to sit on top of a busy gospel song without getting lost in the mix. Not like the life should be taken away from the dynamics of a performance but a little control can keep things on place and even give better definition to a performance.
Two places I see compression over used is as a mastering tool to level out mixes and give a "radio" sound and limiting for mastering in order to produce a loud CD. I still consider the sound of dynamic control on individual tracks or signals to be the best way to control and define a mix. Dynamic control on a two buss does nothing to help keep frequencies in there place. We talk about using EQ to create space for other instruments ...how does dynamic control fit into that? By using dynamic controllers, the space created using EQ can be better restored during a performance keeping a track or mix clean and in place. If you created a space for your bass guitar by cutting some 325Hz on the kick... how are you going to keep the bass guitar from interfering with the fundamentals of the lead vocal in say the 400Hz area? Dynamic control .... keeping things in place. Be more creative with envelope control ... make a snare spank ... find a dynamic controller that has that smooth pillow type push sound for kick. What dynamic controller can handle percussion or drum over heads and open up your drum sound. Are your drum over heads just for cymbals or are they the secret tone grabbers to give drums life?
The more we learn about dynamic control the more creative we can be an engineers. Understanding what dynamic controllers can fix or enhance will give a new found language to your mixes and your talents.