008 – cmprssn two

In the last post we looked at how compressors work and the intended uses for them. Now we will look at a few ways to abuse the effects, and use them in a creative way.

Sidechaining is a commonly used method for both mixing and creative effect. With a side-chained compressor, the trigger signal is different from the affected signal. Traditionally used to dip a bassline out of the way of a kick drum, this can be used for any element within the track, or even with a trigger that is not played through to the master channel. This creates interesting opportunities for allowing elements in similar frequencies to play together, or to create interesting grooves and patterns that may not be replicable simply through arrangement. Sidechaining for mixing purposes is best done with the minimum attack time, and a release that makes sense based on the sounds affected and the tempo and style of the track. When used in a creative way, what sounds good is good.

We can also make use of upward compression, and under-threshold compression. Within Ableton, this can be achieved with the Multiband Compressor, set into a single band if desired. With upward compression, the compressor actually makes sounds that exceed a threshold louder. Depending on the sound, this can be very effective for isolating an element, mixing, or emphasis. Downward compression, below a threshold, is very useful for eliminating some background noise in a sample. This is very similar to the effect of a Gate.

Depending on the analogue type, or digital representation, the output signal can be affected beyond dynamics. This is why certain analogue compressors are sought after studio tools, the LA-2A and 1176 models being prime examples. Invented over fifty years ago, these compressors are still studio staples, due to the way they introduce slight – or overt – distortion into the signal. Modern digital models can also be purchased now, with painstaking effort put in to try to represent the analogue signal path with code.

Lastly, let’s quickly go through overdriving compressors. Certain analogue compressors are still sought after for their particular way of colouring sounds routed through them. Digital models can do this quite well now, of course not getting the analogue feel exactly right, but in a full mix, these slight differences are not usually to be worried about. Ableton’s standard Glue Compressor is a very easy way to start getting some of this effect. With the proper settings for the specific sound, the compressor will start to impart some character and warmth into the sound. Be careful to avoid clipping, in a digital environment this is something to be avoided in all but the most specific circumstances. For compressor overdriving, I often go back to the compressor in Trash 2. When the gain is set very high, the clipping and saturation that happens within the plugin can be very effective musically. The final compressor I use within my process, for distortion purposes, is RoughRider from AudioDamage. This compressor is particularly useful for taming high end with just a touch of warmth.

007 – cmprssn one

Compressors are often totally misunderstood and improperly used by most producers, especially those just starting out. Let’s get into this week going over what they actually do, and then how to best use them. Next week, once we know how they work, we can better understand how to break the rules a little bit.

A compressor makes a sound quieter. That’s all. There is a common misconception that compressors are used to increase volume, and I think this can be largely attributed to the default settings of most VST compressors and a lot of bad information in the online community. Compressors are, however, used in a chain of effects to allow a sound to be perceived with higher volume. Often times, this chain is built into one effect, using a Makeup Gain effect at the end.

Compressors make a sound quieter by “compressing” a signal when it comes in too loud. Originally this was to prevent overloading analogue circuits with an overpowered signal, using an electronic circuit to replace an engineer tasked with dropping a volume slider when a signal got too “hot.” With an analogue circuit or a VST model, “compression” is set as a ratio, from 1:1 through to ∞:1. The first number is the input, or how high the signal goes over a specified threshold, and the second is always 1, as a simplified fraction. 1:1 is no compression, 2:1 reduces the output to 1dB over the threshold for every 2dB the signal is over the threshold. Therefore a 10:1 compression ratio is 1dB over the threshold for each 10dB that the original signal is over by. This decreases the dynamic range, by quieting audio peaks and allowing quieter portions to pass through unaffected. Because of the way our ears perceive audio, and how headroom is processed in our modern digital audio systems, by decreasing the dynamic range of a sound with a compressor, we can actually make it seem louder afterwards. This is another place where confusion sets in.

Before we go over makeup gain, I will briefly cover the two other primary parameters: Attack and Release. These work very differently from the synthesiser envelopes you may be used to. In a compressor, the gain reduction takes place and is released almost instantaneously, from the moment the signal is over the threshold, to the moment it drops back down. Attack and release times tell the compressor how long to wait after the signal goes over and drops back under. So, a 10ms attack does not mean the compressor slowly increases gain reduction over the 10ms. Rather, it will wait for 10ms after the signal goes over the threshold volume, and then immediately reduce the output gain. The same is with the release time.

Makeup gain is the last phase and is the phase that can make the sound louder. Because sounds are made of waves, they have amplitudinal peaks. In a digital environment, clipping happens as soon as a signal exceeds 0dB, and in this instance, clipping is very destructive. Clipping is also a primary tool for distortion, but that is done in a more controlled way. With compression, we can decrease the volume of peaks, thus allowing us to increase the volume of the sound more before the risk of clipping. Then, we can increase the RMS volume while allowing the peaks to stay constant, or even fall from the original reference. RMS is more integral to how humans hear, so compressors will remain to be an incredibly valuable mixing and mastering tool.

Now that we all have a better understanding of compression as a technical tool, come back next week for some ways of using it as a more creative tool.

006 – distorting reality

Continuing on with the theme from last week, let’s dive into some distortion. This is another incredibly important aspect of sound design and mixing.

When used to excess, distortion and saturation can be used to create entirely new sounds. This. for me, is where waveshapers and convolvers come in. These are distortion plugins that allow the user to determine how they want the original sound to be distorted. Ableton has some of this functionality built into Saturator, and other included effects are useful for certain sounds, but I usually go for the sheer control and power of iZotope’s Trash 2. This particular software allows for extraordinarily specific control of what type of distortion you would like, and where you want it. If you haven’t yet explored Trash 2, it will absolutely help you up your distortion game.

Trash allows you to distort your sound with customisable waveshapers, as well as a convolving effect, some resonant filter banks, a few delay models, and a very nice sounding compressor. Ableton has some of these capabilities built into the stock plugins as well. By shaping the way the original wave is processed, the output levels, distortion areas, and added frequency content is all adjustable. Interesting effects can also be created by adjusting any of these parameters in time.

When used to excess, distortion often sounds absolutely terrible. When used in proper amounts, and with the proper setting as adjusted for the sound and the particular effect desired, it will really improve how your tracks sound and feel. What works for you is entirely dependent on style and how you like to work. I do most of my mixing as I compose, so my use of distortion tools is built into my creative process. Experiment with what you like, and try all sorts of plugins for different sounding distortion. Combining multiple effects, whether that is in series or parallel, can be used to make all sorts of wonderfully thick and gritty sounds only you can make. Sounds don’t have to be clean to sound great.

005 – saturated fat

Keeping with the theme of destroying sounds, let’s get into some distortion and saturation goodies. As you may notice by now, a huge part of my music writing process has to do with destroying sounds. With saturation, we can add some lovely (gross) harmonics to sounds to make them punch out of the mix, or just to simply change the character so the sound fits into the track in a smoother, more natural way. Of course, we can also make things stick out in a very non-natural way, depending on the specific sound and what feeling you are going for.

Very simply, saturation is what happens when there is too much audio data for a magnetic tape to store. It is another one of those analogue things we try to emulate in the digital world. It turns out that our ears don’t necessarily like things sounding too “clean.” In digital workspaces, saturation is usually done with a plugin that squishes the waveform slightly and adds some of the randomness associated with the varying magnetic field of a tape. This can, of course, be adjusted from very subtle, to incredibly obvious and over the top. Used sparingly, it functions as a mixing tool. Our ears are more sensitive to slight amplitude changes at higher frequencies, so by adding a little high-end to a low- or mid-range sound, we can actually increase the perceived volume of the primary sound, without changing its amplitude. Used a little more liberally, saturation makes a lot of sounds feel a little bit “warmer.” This is also one of the reasons analogue systems are often preferred by audiophiles and mixing engineers. Analogue systems often impart slight distortions onto sounds, which can make them feel nicer to listen to.

In a digital system, either a little or a lot of saturation can be very important to achieve a complete “feel” for a track. Digital sounds are, by their nature, very clean, and even if we don’t consciously hear this difference, we can sense the subtlety. Just a little bit of saturation, or other types of distortion, as I will go into next week, can make all the difference between a sound that is lost in the mix, to a forward element that is exactly where you want it to be heard.

004 – delay

Delay is great to play with and incredibly useful, especially when you don’t follow any rules, and just go with what sounds awesome. Last week I talked a bit about the Haas Effect, which is a simple way to use a delay to spread out the stereo image and improve the perception of a sound within the mix, but without any echo effects as are usually associated with delay. This week I’ll take a look at some ways to create new sounds and textures with delay.

With many digital delay plugins, there is an option to repitch the delayed sound as the time is changed. This effect is, of course, from the tape delay used in these plugins are modelled on. This effect can be used to change the pitch as the delay time is adjusted throughout the track. I use this to increase tension, as the delay changes speed and thus pitch into a transition within the track. Because the pitch change is based on time, the resultant frequencies are often not harmonic, or in key, with the rest of the track, but for a few beats or bars, and on key sounds in specific places, this technique is incredibly useful for piquing interest and adding flavour to the track.

Some of my favourite plugins are the delay-based Glitch Machines devices. I find myself going back to Hysteresis, Fracture, and Convex, for nearly every track I produce. All of these plugins are fantastic for glitchy delays and aural artefacts that can be changed in time. Convex is particularly useful in this way, with extensive LFO and envelope banks. Not only will they produce delays, but we can also use these plugins to change the pitch of some echoes, use reversed sections of the original audio, change the grain size that is echoed, and of course modify each parameter with an LFO or a repeating envelope. I will use these effects sparingly on main track elements, and heavily on background elements, adding some additional interest and randomness.

As for delay return tracks, I do use one for some builds to increase tension or allow a sound to naturally fall out of the mix, but I primarily set up each delayed sound with its own delay parameters; increasing the complexity of all the subtle and primary delay elements by allowing them to create their own micro-rhythms and strange artefacts created with multiple release times and feedback settings.

003 – dimension

I think something that the brilliant producers of the world do very well is adding a massive level of spaciousness to their mixes. In some clubs and studios, technologies such as Dolby Atmos are making their way into use. What this can do to your experience of the music is incredible, and I think this is a less appreciated way some of the world’s top producers create their own personal dynamic. Here I’ll dive into some of the ways I create space, and how I hear it in other tracks.

With my background atmosphere, I will often use two completely destroyed drum loops, each with separate effects chains. These loops will be panned hard to each side, so there is a difference between the stereo channels to create width. This effect is subtle as I mix this in at a very low level, but it gives me a baseline width and adds body to the track without distracting transients or harmonics.

I will also use a similar effect with my high-frequency elements. If I want to layer hi-hats in any way, usually to thicken the sound and add a chorus-like harmonic, I’ll often set one of these sounds to mono, often the snappier sound, and spread out the thicker sound to the edges. This effect can be achieved in a few different ways. One, with a delay effect set to full wet and very short delay times, but with the left and right times slightly offset. Second, I may also use a left/right split EQ bank with a slightly different frequency curve on each channel. Finally, I may also set a full left and full right channel for the same sound, and ever so slightly detune one or both. Reverb has a place here as well, but this is better used for front-to back space.

Reverb has a place here as well, but this is better suited for front-to back space. One of the tricks I have heard and make use of is sidechain compression of the reverb on a sound, with the peaks of another sound, in doing so adding some groove to the reverb tail and preventing it from taking over the mix. In this way, it is also very useful to use multiple reverb settings or plugins as well, to establish further three-dimensional distance between elements. Pre-delay is an especially powerful parameter to create a perception of distance.

The final element I will mention here is collapsing tracks to mono. Lower frequencies really need to be collapsed to a mono sound, to keep the power and feeling up. Low frequencies really don’t play nice with bad harmonics and benefit greatly from the extra power of being played in sync from both stereo channels. Mono elements also make more apparent the very wide elements, as people notice contrast and difference very acutely.

002 – breaking things

In my mind, one of the primary ways electronic production sets itself apart from acoustic writing is the glutton of choice and possibility when it comes to destroying sounds. I also use certain effects in a way that maybe wasn’t intended. In future posts, I will go into effect and plugin specific techniques, but for now, this will be a brief look at a few of my methods.

The effects can also be added on top of each other in a more subtle way, by using only a little amount of wet signal, bouncing the audio out, and processing again. But that also takes far too much time to be included in a regular workflow.

A simple way to create some noticeable grit and pique some interest is with Ableton’s Warping Modes. This effect can again be consolidated and broken up again. But even without consolidating, it’s still a strong way to break up sounds and change them into something completely different. Simply by changing the bpm by a large amount, and scrolling through the warp modes, adjusting the parameters to extremes will often leave you with a result entirely different from the initial sound. The bpm can be returned to the original later if desired but make sure to consolidate beforehand. This effect is not nearly as powerful when the clip is played back at or very near the original speed.

One effects chain I go back to again and again is made with various instances of a complex delay device, often Hysteresis or Convex from Glitch Machines. This is a great way to add subtle or very apparent glitches and artifacts to a sound, and a great way to fill and spread out a track. A key here is to have a different effect panned hard to each side. This can be a subtle difference or an entirely different patch, it depends on the sound I want and what the track needs. This can be intensive on CPU resources, but with Ableton’s ability to save entire tracks as files, it’s very easy to work around. The processed audio can be flattened and the original track can be saved so adjustments can be made later in the process.

One last thought for this week. Using an analog-modelled delay can create some interesting pitch shifts when the feedback time is adjusted in real time. This is useful for calling attention to sounds echoing out, or simply adding a further dimension into the mix.