Sound Design 101: Synthesizer Basics – Part 1
Having once been the exclusive province of those with the money to afford the latest greatest technological tools of the day, synthesizers are arguably experiencing a golden age of ubiquity. Most DAWs, if not based completely around their included virtual instruments, at the very least include some kind of sampler or subtractive synthesizer plugin. Similarly, thanks to the cost effective evolution of modeling technology, hardware is also plentiful these days. Somehow, despite all this, a lot of electronic musicians seem to be perpetually integrating some new grail plug in into their studio or shopping for the next one when they should be recording. What’s going on here?
Something about the immediacy of the modern DAW has really lent itself to the preset users of the world. It is, in fact, quite easy to confidently saunter through a three chord house music trick with a wobble bass preset and some blistering lead preset or sopping wet pad. Add in your favorite pre-made drum kit, maybe something fancy like “Dubstep Kit 01,” and before you know it, with a bit of competent engineering, you are sitting on something that sounds just like the hits of last year. It isn’t to say this is a bad thing, though the mix might be a little overstuffed with included patch reverbs and delays that were built in to sounds to sell the plugin. It also might not sound horribly original, either, and only takes some other artist who stumbled along the same patches and chord progression to make the creator feel like they need to shop for that next next level plugin or synth. I have heard the problem described as “identakit,” where a number of musicians are cranking out things that ultimately end up betraying how they were created because this preset identifies the plugin/DAW/hardware used.
It does seem as though a majority of programmers, new and experienced, struggle to come to grips with even the most basic elements of the sound design process. Perhaps the buried alchemy of the digital years and their time intensive menu diving oriented programming has taken it’s toll. Perhaps people’s minds simply roll off into thinking about their laundry as soon as words like oscillator and programming get thrown around. Whatever the reason, with so many synths at our fingertips, it’s a wonder there aren’t more sound designers. I spend a lot of time thinking about this when I bump into some other geek at a party and we get to talking shop and cries of “Nerd Alert!” erupt all around us. The most disconcerting fact about this scenario is that I hang out with a number of musicians who use synthesizers.
So, why should one correct themselves on this matter? Improving upon one’s sound design abilities is rooted in the most basic elements of a production. Let’s say your record is not electronic at all, but based in organic instrumentation. It’s a bit of a long shot to think there is some synthesizer or plugin out there that has patches already suited to to way you recorded the cello in a room with your rare tube mic and preamp combo. There are exceptions to the rule, but for the most part, factory patches are designed to sell a synthesizer. There’s a difference between a patch that’s made by a sound designer for the purposes getting that cool new synthesizer off the sales floor and the patch that is going to glue your dance floor smash together.
In designing your own sound, you move past this dilemma. It starts simple: a fairly unmolested filtered but buzzy sawtooth wave to thicken up the bass in the chorus section. Maybe a short decay, fast attack rhythmic plinking noise to help punctuate that hanging guitar part. Whatever it is, these elements become as distinctive to your tracks as the way you record your drums, the board you sum your mixes through, and the individuality of the players themselves. Control over your sound becomes as much of an expression as the parts you write and often informs those parts or even inspires them. Another, less imperative or obvious reason to learn sound design is that, despite the top level appearance of science to be learned; making your own noises is really a lot of fun.
Synthesis is a very tactile and feedback oriented art form. If you approach sound design with a simple and identifiable set of goals, the sweet rush of success is not far away. The most popularized synthesis method, subtractive synthesis, is easy enough to grasp in principle. With basic knowledge of how it works, and a few hints in the right direction, one can be banging out booming basses, rich pads and squelching leads in no time.
As implied by the name, subtractive synthesis, also often referred to as analog synthesis, consists of taking a waveform, or waveforms, generated by an oscillator, or oscillators, and sculpting it by removing or enhancing the harmonic content with filters and modulators in the signal path. We have plans to address other popular synthesis methods, but for now, our focus will be subtractive synthesis. With the amount of freeware available out there today it will be easy for all our readers to follow along with their device and platform of choice. As far as diagrams and examples, I will be using Reason’s Subtractor.
If the name doesn’t make it obvious, Subtractor is your basic 2 oscillator subtractive synthesizer. It has its neat little tricks, but for the most part isn’t going to leave me explaining things that aren’t on your plugin of choice. So, for the purposes of pointing and explaining about different parameters, Subtractor will make for a great point of reference. The information we go through here will be valuable when we have Propellerhead product specialist James Bernard in the house to show off his Reason skills and talk with us about the new version. So, strap in and let’s learn how to make a noise.
The oscillator is the first stop in the synthesis food chain. Oscillators create sound by spinning at a high rate of cycles per second, hence the name. The shape of an oscillator’s wave form is the first stop along the way to determining the harmonic character of the timbre created. Typically, options in a standard subtractive synth will include sine, sawtooth and square waves, not always, but often present is the triangle wave. This is what people mean when they are talking about “waveforms.” There is also usually some kind of “noise” wave available. As for Subtractor, there are a number of other hybrid waveforms available after the initial four, and the second oscillator has a noise component that is meant to be mixed into the signal. Oscillator, as a term, will reappear later on when discussing modulators like LFOs. The difference between the two that is important to understanding the oscillators at the start of a signal path is that these are spinning at such a high rate that they make a noise. Were a square wave oscillator slowed to a certain point there would be obvious gaps in the sound. Were a Sawtooth slowed, you would begin to hear an initial high volume descend to silence, then start back at the top and fall again. So, that’s pretty cool, right?
The basic characteristics of the waveforms mentioned are as follows:
Sawtooth- Literally shaped like a saw blade. The sawtooth is like a series of half triangles joined together end to end. This gives the Sawtooth a rich harmonic character and a biting tone.
Square- The square wave looks like a series of squares spaced apart along a flat line. The gaps in the waveform and the speed with which the wave is presented amount to a hollow, sort of reedy tone.
Sine- The sine wave is like a bunch of S shapes laid on their side and joined together. The smooth fluctuation of the rising and falling tone creates a round and even, soft tone. The sine is often found in bass tones because of its solid performance in lower octaves.
Triangle- The triangle wave splits the difference between the sine wave and the sawtooth. In the end it looks like a series of upturned v’s joined together.
The basic options available when creating sound at the oscillator level (beyond waveforms) are oriented around the pitch and fine tuning of the oscillators themselves. Typically there will be octave adjustments (moving 12 semitones at a time), which allow you to create sounds that have more depth by blending pitches that are spaced further apart. After this, semitone adjustments can be made, which is how 5th leads, 4th leads and the like are created. Next up will be the cents or fine controls, which are for adjusting the fine tuning of your oscillators. Often, what sounds to the ears like a chorused or phasing effect within a single synth patch will actually be two oscillators tuned a few cents in opposite directions to one another. This is the first stop along the path to those lush pads you hear in your head. To some degree, understanding these principles relies on a certain amount of music theory/ fundamental knowledge, but don’t let that stop you from using your nearest synth as a jump off point for acquiring such knowledge through experience.
As you are playing with the oscillator controls, there is one rather unique function on the Subtractor synthesizer: the FM knob. FM refers to frequency modulation synthesis, which is an article in itself. In brief, fm synthesis uses a second oscillator, or operator, to modulate or shape the sounds of the first. In order to hear the FM effects of the subtractor in a non-blended context, you need to turn the oscillator mix dial all the way to the left so that it is only hearing the first oscillator. This will mean that the other oscillator is operating purely as a modulation source. While it would be too deep of a tangent to further delve into the magic that is FM synthesis, pursuit of understanding even this most basic system will result in some sweet 80s bells sounds and some pretty raunchy leads.
The filter is the next stop along the path to synthesizer magic. This is where the squelch happens. The filter is made to allow some frequencies pass through, but not others. Subtractive synthesizers will always include a low pass filter. The low pass filter is so named because it allows the low frequencies to pass through, while cutting out the high ones that are above the frequency you set it to. It is common in a lot of plug ins and some more modern analog synths to see the LP 12 and LP 24. In these the number is referring to the band rejection response rate. This means that the curve of the filter itself will be more gradual in the 12db and more rapid in the 24db. The 24dB falloff rate is a more modern feature, while the 12dB is more common to older synths. The 12dB is your go to when trying to re-create the squelch of a Moog type synthesizer.
So far, I have said a lot about the low pass filter, but it is common to also see high pass, band pass and notch, or band rejection filters in the filter section. Given the logic of the low pass filter, it is obvious that the high pass allows higher frequencies through. To keep it brief, it has the opposite effect of the low pass filter. The band pass filter is essentially a combination of both of the filters previously discussed. Band pass filters do a fixed width filtering out of all sounds above and below a certain distance from the selected frequency. That said, if placed squarely in the middle of the frequency setting, the band pass filter would only allow the sound occupying a certain space in the mid-range of the sound through. That’s simple enough, right?
What remains is the notch filter. The notch filter is more easily addressed as a band rejection filter. Over a smaller range, it actually rejects the selected frequency and lets everything else through. The purpose of this is sometimes to simply remove an unwanted frequency, but it comes to life more obviously when modulation is applied. Don’t worry, we’ll get there, but first we have to discuss the other very important part of the filtering section.
Resonance is an increase of the amplitude of the frequencies immediately surrounding the rolloff/ band rejection point of a filter. This emphasizes the filtering effect in an often desirable way that creates a distorted effect. In simple terms, resonance can best be observed “doing its thing” during a filter sweep. If you push a filter all the way to the top of the spectrum, or the audible and practical spectrum (so let’s say 22.5kHz) and sweep it down over a constant tone from the oscillator, you will hear the filter make a snarly climb to the bottom of the range. Sat in place, it can be used to give some bite to the frequencies in play in a played part. Where resonance really comes alive in the most classic synthesizer sounds is, again, with modulation.
Before we get to the modulators, we should address the two controls specific to subtractor. The first: “Kbd” can logically be assessed as an abbreviation for “Keyboard.” This control appears in other places and on other parameters. In this case, it is a range based parameter that allows for filter frequency to increase/decrease proportionately to the position you are playing on the keyboard. This can be used to compensate for desired filter (etc.) settings that work in one spot of the keyboard, but as a result of the parameter setting chosen, do not work in others.
Filter 2 is the other interesting function that Subtractor brings to the table. Filter 2 is a second 12db low pass type filter. One of the most obvious practical uses for this is being able to rely upon a second low pass filter to control wilder modulations you might set up for filter 1. Where this filter gets interesting, though, is the “Link” control. The Link control will attach filter 2′s frequency control to the position of the frequency control on filter 1. If the first filter is set on HPF (High Pass Filter), this effectively gives you a customizable band pass filter. Similarly, if you are using the notch filter settings, this gives you a LPF (Low Pass Filter) that is moving in concert with it as you adjust the frequency control or effect it with modulators.
Yes, I said modulators again. I suppose it’s finally time to talk about them. However, for today we will take a break and get back to them in part 2. In the meantime, I leave you armed with a working knowledge of two very important components of subtractive synthesis. This should be more than enough to keep you busy for now. Let me know what questions you have and we’ll try to sort it all out.