A little bit about sound
Sound moves in air through a vibration causing a disturbance in air particles. his causes them to move towards each other and eventually transfer the vibration to our eardrum. A sound wave tells us lots of information about the sound itself.
Frequency
In music technology, we use the term frequency to describe how many times something happens in a period of time. Normally, it refers to the number of wave cycles passing a point in 1 second. If 1 wave cycle passes a single point in a second, we say that the frequency is 1 Hertz (Hz).
Processors like EQs and filters use frequency to remove or emphasis treble and bass sounds. Effects with use an LFO (low frequency oscillator) for a continuous change (e.g. a phaser or flanger) have a rate setting which is a frequency control.
PRACTICAL TASK: Analysing frequency content using the EQ and multimeter
Filtering
Filters remove some frequencies from a signal. There are three important types of filter you need to know about:
EQ units are basically a combination of filters and some parametric changes.
The Logic EQ below allows us to make 8 different filter / EQ changes in one interface.
The numbers on the x axis of the EQ represent the human audible frequency range of 20 – 20000Hz.
1000Hz = 1kHz (kilohertz)
Each individual EQ / filter change has different parameters. Parameter is a broad word for something we change. Normally each has 3 parameters – a frequency control, a gain control and a Q / resonance control. We will come back to these!
PRACTICAL TASK: Creating a dance music filter sweep / white noise sweeps
Dynamics
In simple terms, dynamics refer to how loud or quiet something is. It is important to have control of how loud or quiet something is in music technology. We control dynamics using processors like compressors or noise gates.
The decibel (dB) gives a sense of how loud something is. It is a measurement of the sound pressure level (SPL). It has what we call a logarithmic scale because the actual sound pressure needed to make something seem twice as loud is not twice as much, it’s much more. More on this later!
Compression
A compressor is used to make the volume of a track more consistent. It is essentially an automatic volume knob that turns the volume down when the signal gets too loud.
We stay it reduces the dynamic range because it reduces the difference between the loudest and quietest sounds in a signal.
It does this by pushing the peaks of a signal down, and then raising the overall volume using make up gain.
In the diagram above, the peaks have been pushed down so the overall volume of the signal is more consistent. We then raise the overall volume (make up gain) which means the signal overall is louder than the uncompressed signal, and the dynamic range is narrower because there are fewer loud bits that jump out.
We use compression extensively in music technology. Parts that normally need the most compression are vocals, bass guitar, kick and snare drum. The parameters you need to know are:
| Threshold | The volume at which the compressor will start to compress the signal (a minus number measured in dB) |
| Ratio | Once the signal has gone above the threshold (and is therefore compressed), the ratio tells the compressor how much to reduce the peaks by. If the ratio is 8:1, an 8db increase in signal input volume will produce a 1dB output volume. |
| Make up gain | The amount the overall volume is turned up after compression to compensate from the reduction in the volume of the peaks. |
A limiter is an extreme compressor (with a ratio of infinity:1). Limiters are normally used on the output of a project to control the odd wayward peak or in mastering.
Compression can also be used creatively to create ducking / pumping effects in EDM (electronic dance music).
An easy way of thinking about a compressor is that it behaves in exactly the same way as your mum when you’re playing music too loud.
You play your music.
Your mum listens downstairs until the music reaches a point that is too loud for her (the threshold).
She comes up the stairs and turns the music down. The time this takes is the attack time.
The amount your mum turns the music down by is the ratio.
She goes back downstairs. You decide you can’t hear the music any more as the next track you’re listening to is much quieter.
You turn the music back up to where it was. The time this takes is the release time.
PRACTICAL TASK: Compressing kick / snare / bass
Gating
A noise gate silences a signal when it is below the threshold you set. Noise gates are really useful for removing noise between phrases. They can’t remove noise whilst the signal is playing.
Noise gates are really useful when recording drums to make the kick and snare more isolated. This is done by cutting out the spill from the other elements of the drum kit on e.g the snare track.
You can also use a noise gate to silence vocal noise or distorted electric guitar buzz between phrases.
| Threshold | The volume above which the signal can pass through the noise gate (a minus number measured in dB) |
| Reduction | How much the gate reduces the signal volume by (-100dB = silent, 0dB = not at all) |
| Attack | How quickly the gate opens (in ms) |
| Hold | How long the gate stays open when the signal goes under the threshold (in ms) |
| Release | How long the gate takes to close once the signal is under the threshold and the hold time is complete (in ms) |
Attack / release times of 0ms will normally result in a click.
An easy way of thinking about a noise gate is as a fountain with a hinged metal cover over the top:
The threshold is how high the pipe is that the water comes out of; the attack time is how long the metal cover takes to open when water hits it
The hold time is how long the metal cover stays open when the water drops below the top of the pipe; the release time is how long the metal cover takes to close.
Reverb
Reverb sets a sound into a space. Reverb can either be a natural product of a space you are recording in (e.g. room / hall reverb), it can be added artificially from another space (e.g. chamber), produced using artificial mechanical means (e.g. plate / spring) or digitally recreated (e.g. digital / convolution).
Music technology students often overuse reverb in their non-examined assessment work; modern commercial pop music is much drier than it was say, in the 1980s.
We can think of reverb in three ‘parts’ which happen after the direct sound.
First there is the pre-delay (the gap between the direct sound and the first early reflection)
Then the early reflections (single reflections from nearby surfaces)
Then the reverb tail (a wash of multiple reflections that, gradually gets quieter)
| Reverb time (RT60) | How long it takes the reverb to decay (1s – a little reverb; 2s – a normal amount of reverb; 3s – a lot of reverb) |
| Pre-delay | The gap in time between the dry signal and the reverb. |
| Wet / dry | The balance of the effected vs unaffected signal (sometimes called ‘mix’) |
PRACTICAL TASK: Creating contrasting ambience with Space Designer and ChromaVerb
Delay
A delay repeats a sound after you’ve heard it. Delays are used extensively in music technology as effects in their own right and to create e.g. reverb and modulation effects.
Historically, delays were created in analogue ways (tape, bucket brigade chips) up until the 1980s and then digital delay units made inroads into the market, followed by delay plugins on computers.
The parameters you need to know about are:
| Delay time | How long there is between the original signal and the repeat (often timed to a musical note) |
| Feedback | How much of the delayed signal is fed back into the input to create a series of delays |
| Dry / wet | How loud the delay is compared to the original signal |
| Filtering | Whether the delays have HFs / LFs removed |
| Stereo | Do the delays stay mono or do they go from e.g. left to right |
PRACTICAL TASK: Create a subtle delay with the Logic Tape Delay, and a complex rhythmic delay with Delay Designer
Distortion
Distortion is created by turning up the volume of a sound so that it is too loud. This introduces clipping into the signal. This is where the amplifier / system cannot get any louder and therefore squares or rounds off the top of the signal peaks.
This happens in lots of different places in music technology; guitar pedals intentionally clip the signal to give an overdriven or fuzz effect. You can get tape emulator plugins for your DAW which copy how tape soft clips / saturates the signal when it gets too loud (a desirable kind of distortion).
Turning up the gain gives more distortion. This adds lots of harmonics to the signal.
PRACTICAL TASK: Adding harmonics using distortion / reamping a guitar part
Modulation
You might be used to the word ‘modulation’ meaning a change of key in traditional music if you’ve studied GCSE / ABSRM lessons.
In music technology, modulation has a broader definition of a change over time.
Vibrato modulates pitch over time.
Tremolo modulates volume over time.
In the screenshot below from the Logic Tremolo effect, the volume of the signal turns up and down according to the rate.
Chorus and flanger are both modulated delay effects. They combine the original signal (the dry signal) with a signal that is moving in and out of time by a little bit (the delay is modulated – the wet signal). This creates a sense of detuning and has an impact on the frequency content of both signals as they interact with each other. Chorus sounds shimmery / layered. Flanger sounds whoosy / jet plane like. A phaser is a bit like a flanger and sounds similar – we’ll come back to this.
PRACTICAL TASK: Applying and comparing chorus / flanger / phaser effects. Making your own chorus.
OTHER FX AND PROCESSORS TO KNOW ABOUT
| Vocoder | A robot voice sound effect where the voice is used to shape a synth sound (think Daft Punk / Pet Shop Boys synthpop) |
| Talkbox | A tube goes in the guitarists mouth; they shape the sound by changing their mouth position (think Bon Jovi – Living on a Prayer) |
| Bitcrusher | Lowers the bit depth to simulate digital distortion. Sounds robot like / chiptune / crunchy and grainy (think La Roux grainy synth parts) |
| Wah wah | Onomatopoeic – sounds like it is said; a footpedal is moved up and down to shape a guitar sound (think Jimi Hendrix) |
| Autotune / pitch correction | Snaps a vocal / instrument pitch to a grid or scale; if overused, can sound robotic / R&B effect (think Cher, T-Pain, Kanye West – if used subtly, corrects small tuning errors and is difficult to hear |
| Ring modulation | Robotic sound with lots of odd shimmery harmonics – sounds a bit like singing through a fan (think Black Sabbath – Iron Man) |
PRACTICAL TASK: Listening examples for each other FX / processor; applying them in Logic
Inserts and sends
Inserts and sends are two different ways of adding effects to your tracks in Logic. You also see inserts and sends on real mixing desks.
An insert puts a copy of a plug-in on a single track; you ’insert’ the effect into the signal flow. With a send, you take the track and route a (partial) copy of it to an auxiliary channel, and put your effects on the new channel. This is much more efficient for CPU usage.
In general, reverb and delay are best used as send effects. This means you can also use them to blend different instruments together. You might want to use an insert if the effect is only going to apply to a single track, or you want to further process the track; the order of inserts makes a difference.
PRACTICAL TASK: Adding reverb as a send and as an insert
PRACTICAL TASK: Tightening up the tuning of a vocal part / creating an R&B effect
Sampling
Sampling is when you take a part of a song, single note or sound and reuse it in another context.
It is common to use a sampler to either record, manipulate or playback one of these pieces of audio material (or any combination of the three).
Originally in the 1960s, samplers used tape loops. To change the pitch on a tape-based recording, the tape was played faster or slower. However, unfortunately, tapes were subject to hiss, wow and flutter, and degradation.
In the 1980s, digital samplers were developed. These stored the audio signal as digital data – readings of the volume of the waveform. We call these readings, confusingly, samples. The speed we take these readings is the sample rate; the accuracy of these measurements is given by the bit depth.
Because 80s samplers didn’t have much storage space, musicians had to keep the sample rate and the bit depth low. This made the samples muffled and grainy / noisy.
Now sampling is mostly done on a computer. This provides quicker, easier and more detailed editing. It is important to cut a sample at a zero crossing point to avoid it clicking.
Common sample manipulation techniques include triggering, reversing, gapping and stuttering, beat slicing and pitch shifting / time stretching.
PRACTICAL TASK: Sample editing using the Quick Sampler
Synthesis
A synthesiser is an electronic sound generator capable of creating and manipulating synthetic sounds.
Analogue synthesis uses an electrical voltage to generate a signal which is then shaped. Digital synthesis can use computer technology to model other synthesisers and generate sounds. FM synthesis (frequency modulation) is a type of digital synthesis.
It has become common to use synthesisers as DAW plug-ins, but the sounds, warmth and authenticity of vintage analogue equipment are highly regarded by many.
Subtractive synthesisers start with a harmonically rich waveform & use filters to remove frequencies. Additive synthesisers layer up simple waveforms (e.g. sine waves) to produce a more complex wave. Other types of synthesisers include FM, granular, wavetable and sample-based.
A synthesiser’s polyphony tells us how many notes it can play simultaneously. A monophonic synthesiser can only play one note at once.
Glide or portamento allows a synth (often monophonic) to slide between two notes.
Synthesiser modules
Waveforms
All of the other waveforms are made from layered sine waves. We’ll come back to this!
White noise
A noise generator creates white noise; a random signal consisting of all frequencies at an equal amplitude. It is used to simulate wind or percussive sounds like cymbals, and can also be filtered to create a sweeping effect.
PRACTICAL TASK: Creating a resonant white noise sweep
Envelopes and LFOs
Envelopes are used to shape the sound by altering the attack, decay, sustain and release and thus how the note starts, sustains and finishes. They control a parameter on one of the modules, for example volume (amplifier), cut off frequency (filter) or pitch (oscillator).
Attack – the amount of time taken to reach the maximum value for the parameter.
Decay – the amount of time taken to fall from the maximum value of the parameter to the sustain level.
Sustain – the level at which the parameter stays at whilst the key is being pressed.
Release – the amount of time take an to fall from the sustain level to 0 when the key is released.
Like an envelope, an LFO (low frequency oscillator) is a control signal used to alter a parameter over time. Most synthesisers can use an LFO to control a parameter on one of the modules. Whereas an envelope happens only once, an LFO creates a constant cyclic change. In the diagram below, an LFO is being used to change the pitch over time according to a sine wave.
PRACTICAL TASK: Creating different envelope / LFO effects and envelope / LFO bingo
The recording eras
We will study the history associated with each recording era more closely in chapter 4. However, here is a brief overview of the five recording eras we divide developments in music technology into. These will form the basis for your listening when we look at a new piece of music. This is a helpful summary for your revision that it will be worth revisiting in the future.
| Era | Characteristics |
| Direct to tape mono recording c. 1930 – 1963 | Tape hiss with poor signal-to-noise ratioIndistinct balance Lack of EQ clarity; muddiness / mid-heavy / lack of HFs Limited tracks meaning the drums often fall back and the kick and snare don’t tend to drive the mix Crackle, as some recordings might only have been released on vinyl. |
| Early analogue multitrack recording c. 1964 – 1969 | Hiss built up because of layering and bouncing down Still limited tracks so some instruments may be masked in the balanced of the track More FX & experimentation as more tracks The panning on early stereo releases (which were regarded as inferior) was polarised Generally lacking in HFs / LFs (mid-heavy, but less so than direct-to-tape mono) |
| Large scale analogue multitrack recording c. 1969 – 1995 | Increased clarity (better frequency response and dynamics for individual instruments) because of more tracks being available Multiple mics and experimentation; track count didn’t matter as much Dolby helped reduce hiss because of many tracksExtensive layering and overdubbing Some electronic instruments e.g. synths/drum machines so use of DI common |
| Digital recording and sequencing c. 1980 – present day | Less hiss; brighter mixes with better high frequency response Repetitive loops which often go on throughout a section or entire piece Sequencing meaning layers of synths and drum machines were common Digital sampling, but with lo-fi samples because of the limited disk space; quality improved in the 90s |
| DAWs and emerging technologies c. 1996 to present day | Flawless performances; these might sound over-processed but it’s possible to achieve a more natural sound too Audio quantise, flex time and pitch correction with unlimited editing Lots of layering; there might be endless guitar overdubs or 10s of vocal takes to give a thick sound Low and high frequency heavy masters |
PRACTICAL TASK: How could you mimic / emulate the features of previous recording eras using modern production techniques? Have a go on Logic.