Ways To Generate Distortion
By Roey Izhaki
Mixing is a dynamic art. Mixing trends change through the years, and throughout past decades the extent to which and the way we use different tools has changed as well. There is little argument that distortion is the tool that has gained the most popularity in recent years. Distortion, in that specific mixing context, is not exactly the screaming pedal we use with guitars, but the more subtle harmonic distortion, in its various forms.
There are two main reasons why distortion has become so widely used. First, pop mixes nowadays tend to be more aggressive. This applies to both the analog and digital domains, where distortion is added to various degrees depending on the production. Strictly speaking, the use of distortion is not limited to rock or metal productions – even genres such as dance, chart-pop and trip-hop might involve a certain degree of distortion. Second, distortion is used to compensate for what can be described as the ‘boringly clean’ digital sound – the inherent distortion of analog equipment and tape media is not an asset of digital audio. Engineers use distortion to add some edge to digital audio by introducing some degree of ‘appealing error’.
WAYS TO GENERATE DISTORTION
A boost on the gain control, affecting the signal early in the signal path, can overload successive components. In the analog domain we usually speak first of overloading, which mostly adds subtle distortion, and only then saturation, which can produce drastic distortion. The nature of solid-state components is that they provide increasing distortion with gain. Tube equipment tends to produce a very appealing distortion up to a specific point, where the system seems to ‘break’ and produce very undesired clicks. Tapes can be overloaded using the same principle by boosting the input signal. Bipolar transistors, FET, tube and tape distortion all have different qualities as they all produce different harmonic content. Also, the more each system overloads, the more compression occurs, which makes the overall effect even more appealing.
Of all the types of distortion, digital clipping is probably the least appealing. The harsh limiting of audio exceeding the 0 dB threshold on a digital system can produce extremely unpleasant sound, especially if the added content aliases. There are also different types of digital clipping, where differences are dependent on what happens to signals that exceed the system limit. On most floating-point systems these signals are trimmed to the highest possible value (simply hard-limiting). However, different integer notations can produce different results, where exceeding signals can alias around the highest sample value; they can be displaced to the bottom of the value range or displaced across the zero line.
Generally speaking, clipping tends to produce the strongest type of distortion. Some plugins try to imitate the clipping sound of analog devices by introducing additional processing along with the basic digital clipping. The clip distortion in Figure 24.1 is one of them. It is worth noting the Mix control, which lets us blend the undistorted and distorted sounds.
Short attack and release on dynamic range processors
A short attack and release can act within the cycle of low frequencies, thus altering the shape of the waveform and producing distortion. This type of distortion is not reserved solely for compressors and can be produced by short time constants on any dynamic range processor. We can use this type of distortion to add some warmth and definition to instruments with low-frequency content, notably basses.
A wave shaper is a pure implementation of a transfer curve. Essentially, the signal passes through a transfer characteristics function over which we have some control. This is somewhat similar to a dynamic range processor with no time constants and an unrestricted transfer curve. Wave shapers are not very common and might be a sub-facility within a different effect. However, they can produce extremely interesting results that can be used in subtle amounts for gentle enhancements or in a drastic form for creative effect.
Early digital samplers that emerged around the early 1980s had low specs compared to those used today – in addition to low sample rates, they were designed around 8-bit (and later 12-bit) samples. Even now, the 8-bit sound is much sought after in genres such as hip-hop, where many find the lo-fi sound of drum samples appealing.
When we discussed software mixers in Chapter 10, we discussed the importance of dither, which rectifies the distortion caused by bit reduction. Unsurprisingly, like many other areas in mixing where we are after the ‘technically wrong’ and the less precise sound, bit reduction can also have an appealing effect. Essentially, a bit reduction process simply reduces the bit depth of digital audio. Although within the audio sequencer the audio is still represented by 32-bit float numbers, sample values are quantized to the steps of the target bit depth. For example, reduction to 1 bit would produce 32-bit float values of 1.0 and – 1.0 only (in practice, however, most processors also output 0.0 as a possible sample value). This process produces quantization distortion, where the lower the target bit depth is, the more the distortion. Clearly, we wish to keep this distortion, so no dither is applied. Just like with drum samples, bit reduction can be used to give a lo-fi sense to various instruments. It can also be used as a creative effect.
A famous mixing practice in the analog domain is called re-amping – feeding a guitar recording during mixdown back into a guitar amp so a different sound texture and distortion can be created. Mostly, the signal sent to the amps during mixdown is the direct recording (before any pedals or amp). Many plugins nowadays are designed to imitate the sound of classic guitar cabinets, and usually these plugins also include other guitar processors such as tremolos, compressors, gates, echo engines and guitar reverbs. When there is a direct recording of a guitar, we can easily choose the final guitar sound during mixdown (together with the cabinet recording; perhaps replacing it).
Even before the digital age, re-amping was not limited to guitars. Sometimes engineers sent vocals, drums or other tracks into a guitar amplifier (or guitar pedals), using the amp as a distortion or effects unit. Using the same principle, an amp simulator plugin can be used to distort any type of material. Like any other type of distortion, subtle settings can be useful for gentle enhancements and more trashy settings for more drastic results.
Dedicated units or plugins
As distortion becomes more popular, more dedicated units and even more plugins are appearing in the market. The Culture Vulture by Thermionic Culture is one respected hardware unit. The Soundtoys Decapitator, shown in Figure 24.5, is an analog saturation modeler plugin that can produce a variety of distortion options.
The above is an excerpt from Roey Izhaki’s book Mixing Audio, 2e. Roey Izhaki has been involved with mixing since 1992. He is an academic lecturer in the field of audio engineering and gives mixing seminars across Europe at various schools and exhibitions. He is currently lecturing in the Audio Engineering department at SAE Institute, London.