Snoman R. Dance Music Manual: Tools, Toys, and Techniques

Подождите немного. Документ загружается.

PART 1

Technology and Theory

until the transient of the kick sounds right for the mix. This, however, is only to

acquire a basic kick sound, and it’s now open to tweaking with all the param-

eters you have at your disposal, which incidentally are many more than the

humble 909 or 808 have to offer.

Firstly, as we’ve already touched upon the transient where we attain most

of the information about a sound, adjusting the ﬁ lter cut-off and resonance

of the square wave will dramatically change the entire character of the kick.

A high-resonance setting will produce a kick with a more analogue character,

while increasing the ampliﬁ er’s decay will produce a kick that sounds quite

‘ boxy’. Increasing the ﬁ lter’s cut-off will result in a more natural sounding

kick, while increasing the pulse width will create a more open, hollow timbre.

Additionally, the oscillator producing the sine wave, can also be affected with

the synthesis parameters on offer. Using the pitch and pitch envelope param-

eters you can adjust how the pitch reacts on the sine wave, but more impor-

tantly, you can determine how ‘boomy ’ the kick is through the pitch decay.

In this context, this will set the time that it takes to drop from the maximum

pitch change to the sine wave’s normal pitch. Thus, by increasing this, the pitch

of the sine wave doesn’t fall as quickly, permitting the timbre to continue for

longer creating a ‘boomy ’ feel. Similarly, decreasing it will shorten its length

making it appear snappier. More interestingly, though, if you can adjust the

properties of the envelope’s decay slope you can use it to produce kicks that are

fatter or have a smacking/slapping texture.

If the decay’s slope remains linear, the sound will die in a linear fashion pro-

ducing the characteristic 909 kick sound. However, if a convex slope is used

in its place, as the pitch decays it will ‘bow ’ the pitch at a number of fre-

quencies, which results in a kick that’s more ‘rounded’ and much fatter. On

the other hand, if the slope is concave, the pitch will curve ‘inwards ’ during

the decay period producing the sucking/smacking timbre similar to the E-mu

Drumulator. By increasing the length of the pitch decay further these effects

can be drawn out producing kicks that are suited towards all genres of dance.

It should be noted here, however, that not all synthesizers allow you to edit

the envelope’s slope in such a manner but some software samplers (such as

Steinberg’s HALion) will allow you to modify the slope of a sample. Thus, if

you were to sample a kick with a lengthy decay, you could then modulate the

various stages of the envelope.

Alternatively, if the synthesizer is quite substantial it may allow you to modu-

late not only the sine wave but certain aspects of the envelope with itself.

Fundamentally this means that the pitch envelope modulates not only the sine

wave but also its own parameters. Using this you can set the synthesizer’s mod-

ulation destination to affect the oscillator’s pitch and its own decay parameter

by a negative or positive amount, which results in the decay becoming convex

or concave, respectively. This effect is often referred to as ‘recursive ’ modula-

tion, but as mentioned, this is only possible on the more adept synthesizers.

Nevertheless, whether recursive modulation is available or not, the key, at this

Programming Theory

CHAPTER 4

stage, is to experiment with different variations of pitch decay, the ﬁ lter sec-

tion on the square wave and both oscillators. For instance, replacing the sine

wave with a square produces a ‘clunky’ sound, while replacing it with a triangle

produces a sound similar to the Simmons SDS-5.

While these methods will produce a kick that can be modiﬁ ed to suit all genres,

for hip-hop you can sometimes glean better results using a self-oscillating

ﬁ lter and a noise gate. If you push the resonance up until it breaks into self-

oscillation it will produce a pure sine wave. This is often purer than the oscilla-

tors themselves and you can use this to produce a deep tone that’s suitable for

use as a kick in the track (usually 40 Hz). If you then programme a 4/4 loop in

a MIDI sequencer and feed the results into a noise gate it can be used as an EG.

While playing the loop, lower the threshold so that only the peaks of the wave

are breaching, set the attack to zero and use the release to control the ‘kicks’

delay. This kick can be modiﬁ ed further by adjusting the hold time, as this will

allow more of the peak through before entering the release stage.

Once the basic kick element is down it will most likely beneﬁ t from some com-

pression but the settings to use will depend on the genre of music. Typically,

house, techno and trance will beneﬁ t from the compressor using a fast attack

so that the transient is crushed by the compression. This produces the emblem-

atic club kick while setting the compressor so that the attack misses the tran-

sient but grips the decay stage allowing it to be raised in gain to produce the

characteristic hip-hop, big beat or drum ‘n’ bass timbre.

SNARE DRUMS

The snare drum in most dance music is derived (somewhat unsurprisingly)

from the TR909, or, in the case of house, the E-mu Drumulator or the Roland

SDS-5. All these, however, were synthesized in much the same way by using a

triangle oscillator mixed in with pink or white noise that was treated to posi-

tive pitch movements. This can, of course, be emulated in any synthesizer by

selecting a triangle wave for the ﬁ rst oscillator and using either pink or white

noise for the second. The choice between whether to use pink or white noise

depends on the overall effect you wish to achieve but, by and large, pink noise

is used for house, lo-ﬁ and ambient snares while white is used for drum ‘n’

bass, techno, garage, trance and big beat. This is simply because pink noise

contains more low-frequency content and energy than white and hence

produces a thicker, wider sounding timbre.

To produce the initial snare sound much of the low-frequency content will

need to be removed, so it’s sensible to employ a high-pass, band-pass ﬁ lter

or notch ﬁ lter depending on the type of sound you require. Notching out the

middle frequencies will create a clean snare sound that’s commonly used in

breakbeat while a band pass will add crispness to the timbre making it suitable

for techno. Alternatively, using a high-pass ﬁ lter with a medium resonance set-

ting will create the house ‘thunk’ timbre. As with the kick drum, snares need

to start immediately on key press and remain fairly short, so the amp’s EG will

PART 1

Technology and Theory

need setting to a zero attack, sustain and release while the decay can be used to

control the length of the snare itself. In some cases, if it’s possible in the syn-

thesizer, it’s prudent to employ a different amp EG for both the noise and the

triangle wave. This way the triangle wave can be kept quite short and swift by

using a fast decay while the noise can be made to ring a little further by increas-

ing its decay parameter. The more this is increased, the more atmospheric snare

will appear allowing you to move from the typical techno snare, through big

beat and trance before ﬁ nally arriving at ambient. What’s more, this approach

may also allow you to use a convex envelope on the noise to produce a smack-

ing timbre similar to the Nine Inch Nails (NIN) ‘Closer’ snare.

If two amp EGs are not available, small amounts of reverb can help to lengthen

the timbre, and if this is followed with a noise gate with a fast attack and short

hold time, the decay can be used to control the amount of ambience in the

loop. Even if artiﬁ cial ambience isn’t required, employing a gate can be particu-

larly important when programming house, drum ‘n’ bass and hip-hop loops as

the snare is often cut short in these genres to produce a more dynamic loop.

Additionally, for drum ‘n’ bass the snare can then be pitched further up the

keyboard to produce the characteristic bright ‘snap’.

This initial snare can be further modiﬁ ed using a pitch envelope to modulate

both oscillators and can be applied either positive or negative depending on

the genre of music. Most usually, small amounts of positive pitch modulation

are applied to force the pitch downwards as it plays but some house tracks will

employ a negative envelope to create a snare that exhibits a ‘sucking’ nature

resulting in a thwacking sound. If you decide to use this technique, how-

ever, it’s often worth removing the transient of the snare in a wave editor and

replacing it with one that uses positive pitch modulation.

The two combined then produce a sound that has a good solid strike but

decays upwards in pitch at the end of the hit. If this isn’t possible in the syn-

thesizer/wave editor then a viable alternative is to sweep the pitch from low to

high with a sawtooth or sine LFO (provided that the saw starts low and moves

high) set to a fast rate or programme a series of control change (CC) messages

to sweep it from the sequencer. Once this is accomplished, small amounts of

compression set so that only the decay is squashed (i.e. slow attack) will help

to bring it up in volume so that it doesn’t disappear into the rest of the mix.

HI-HATS

Hi-hats can be synthesized in a number of ways depending on the type of

sound you require but in the ‘original’ drum machines they were created with

nothing more than ﬁ ltered white noise. This can be accomplished in most syn-

thesizers by selecting white noise as an oscillator and setting the ﬁ lter envelope

to a fast attack, sustain and release with a medium-to-short decay. Finally, set

the ﬁ lter to a high pass and use it roll off any frequencies that are too low to

create a high hat. The length of the decay parameter will determine whether

the hi-hat is open or closed (open hats have a longer decay period).

Programming Theory

CHAPTER 4

While this is the best way to produce a typical analogue hi-hat sound it can

sound rather cheap and nasty on some synthesizers, and even if it produces the

timbre, it can still appear quite dreary. As a result, a much better approach is to

use either ring or FM as this produces a hi-hat that sounds sparkling and ani-

mated, helping to add some energy to the music. Ring modulation is possibly

the easiest solution of the two simply consisting of modulating a high-pitched

triangle wave with a lower pitched triangle. FM consists of modulating a

square or sine wave with a high-pitched triangle oscillator. The result is a high-

frequency noise waveform that can then be modiﬁ ed with a volume envelope

set to a zero attack, sustain and release with a short-to-medium decay. If FM

is used and the modulator source is modiﬁ ed with a pitch envelope and the

amount of FM is increased or reduced, the resulting waveform can take on

more interesting properties, so it’s worthwhile experimenting with both these

parameters (if available). Once this basic timbre is constructed, shortening the

decay creates a closed hi-hat while lengthening it will produce an open hat.

Similarly, it’s also worth experimenting by changing the decay slope to convex

or concave to produce fatter or thinner sounding hats.

Notably, unlike most percussive instruments, compression should not be used

on hi-hats as all but the best compressors will reduce the higher frequencies

even if the attack is set so that it skips by the attack stage. This obviously results

in a dull sounding top end of a mix, so any form of dynamic restriction should

be avoided on high-frequency sounds, which include shakers, cymbals, cow-

bells, claves and claps.

SHAKERS

Shakers are constructed in a fashion similar to the high hats. That is, they

are created from white noise with a short attack, sustain and release with a

medium decay on the ampliﬁ er and ﬁ lter envelope. A high-pass ﬁ lter is then

used to remove any low-end artefacts to produce a timbre consisting entirely

of higher frequencies. Once you have the basic ‘hi-hat’ timbre, the decay can be

lengthened to produce a longer sound, which is then treated to an LFO modu-

lating the high-pass ﬁ lter to produce some movement. The waveform, rate and

depth of the LFO depend entirely on the overall sound you want to produce,

but as a general starting point, a sine wave LFO with a fast rate and medium

depth produces the typical ‘shaker’ timbre. Again once this initial timbre has

been constructed, like all drum sounds, it’s worth experimenting by changing

the envelope’s attack and decay slope from linear to convex or concave.

CYMBALS

Again, cymbals are created in a fashion similar to hi-hats and shakers as they’re

constructed from noise, but rather than use the noise from an oscillator, it’s

commonly generated from ring or FM. Using two square waves played high on

the keyboard, detune them so that they’re approximately two octaves apart and

PART 1

Technology and Theory

set the amp’s EG to a fast attack with no release or sustain and a medium decay

(in fact similar to hi-hats and shakers). Once the tone is shaped, both need to

be fed into a ring or cross modulator to produce the typical analogue cymbal

noise timbre.

The attack of the cymbal is particularly important, so it may be worth synthe-

sizing a transient to drop over the top or using a ﬁ lter envelope to produce

more of an initial crash. The ﬁ lter envelope is set to a fast attack, sustain and

release with a medium-to-short decay as this produces an initial ‘hit’ but you

can synthesize an additional transient using an oscillator that produces pink

noise with the same ﬁ lter settings as previously mentioned. If ring modulation

is not available on the synthesizer, a similar sound can be created using FM.

This consists of modulating a high-pitched square wave with another, lower

pitched, square or a triangle wave. As with the FM used to create high hats,

by modifying the source with pitch modulation or increasing/decreasing the

amount of FM you can create a number of different crash cymbals.

CLAPS

Claps are perhaps the most difﬁ cult ‘percussive ’ element to synthesize because

they consist of a large number of ‘snaps’ all played in a rapid, sometimes pitch

shifting, sequence. Although in the interests of theory we’ll look at how they’re

created, generally speaking you’re much better off recording yourself clapping

(remember to run the mic through a compressor ﬁ rst, though, as the transient

will often clip a recorder) and treating them to a chorus or harmonizing effect

or, simpler still, sampling some from a CD or record.

Generally, claps are created from white noise passed through a high-pass ﬁ lter.

The ﬁ lter and amp EGs (as should be obvious by now) are set to a fast attack

with no release or sustain and a decay set to suit the timbre you require –

midway is a good starting point. The ﬁ lter cut-off, however, often beneﬁ ts from

being augmented by a sawtooth LFO set to a very fast rate and maximum depth

to produce a ‘snapping’ type timbre. This produces the basic tone but you’ll also

need to use an arpeggiator to create the successive snaps to follow. For this, pro-

gramme a series of staccato notes into an MIDI sequencer and use these to trig-

ger the arpeggiator set to one octave or less so that it constantly repeats the same

notes in a fast succession. These can be pitched downwards, if required, using

a pitch envelope set to a positive depth on the oscillator but it must be set so

that it doesn’t retrigger on each individual note otherwise the timbre turns into

mushy pitch-shifted clutter. Although claps are difﬁ cult to synthesize it is often

worth the effort required, however, as adjusting the ﬁ lter section and/or the decay

slopes of the amp and ﬁ lter’s EG opens up a whole new realm of clap timbres.

COWBELLS

Fundamentally, cowbells are quite easy to synthesize and can be constructed in

a number of ways. You can use two square oscillators or a triangle and a square

Programming Theory

CHAPTER 4

depending on the sound you require. If you require a sound with more body

then it’s best to use two square waves, but if you prefer a brighter sound then a

square mixed with a triangle will produce better results.

For a cowbell with more body, set two oscillators to a square wave and detune

them so that the ﬁ rst square plays at C Ⲇ5 (554 Hz) while the other plays at G Ⲇ 5

(830 Hz). Follow this by setting the amp envelope to a fast attack with no release

or sustain and a very short decay. This resulting tone is then fed into a band-pass

ﬁ lter which can be used to shape the overall colour of the sound. Alternatively, if

you want to create a cowbell that exhibits a brighter colour to sit near the top of

a mix it’s preferable to use just one square wave mixed with a triangle wave. The

frequency of the square should be set around 550 Hz and the triangle should be

detuned so that it sits anywhere from half an octave to an octave from the square,

depending on the timbre you require. Both these are then ring modulated and

the result is fed through a high-pass ﬁ lter allowing you to remove the lower fre-

quencies introduced by the ring modulation. Once these basic timbres are cre-

ated, the amp’s EG can be lengthened or shortened to suit the current rhythm.

CONGAS

Congas are constructed from two oscillators with a dash of FM to produce a

‘ clunky’ timbre. These can be easily constructed in any synth (that features FM)

by setting the ﬁ rst oscillator to a sine wave and the second to any noise wave-

form. The sine wave amp’s EG is set to a very fast attack and decay with no

release or sustain to produce a click which is then used as FM for the noise

waveform. The noise waveforms amp EG needs to be set to a fast attack with

no release or sustain and the decay set to taste.

There is little need to use a ﬁ lter on the resulting sound, but if it seems to have

too much bottom or top end then use a low-pass or high-pass ﬁ lter to remove

the upper or lower frequencies consecutively. In fact, by employing a high-pass

ﬁ lter and reducing it slowly you can create muted congas while adjusting the

noise amp’s decay slope to convex or concave can produce the typical slapped

congas that are sometimes used in-house.

TAMBOURINES

Tambourines, like claps, are difﬁ cult to synthesize as they essentially consist of

a number of hi-hats with a short decay each occurring one after the other in a

rapid succession which is passed through a band-pass ﬁ lter. This means that

they can initially be constructed by using white noise, FM or ring modulation

in the same manner as high hats. Once this basic timbre is down, the tone is

augmented with a sawtooth LFO set to a fast rate and full depth. After this,

you’ll need to programme a series of staccato notes to trigger the synthesizer’s

arpeggiator to create a series of successive hits. The decay parameter of the

amp’s EG can then be used to manipulate the tambourines, character, while

the results are fed into a band-pass ﬁ lter which can be used to shape the type of

PART 1

Technology and Theory

100

tambourine being used. Typically, wide band-pass settings will recreate a tam-

bourine with a large tympanic membrane and thinner settings will recreate a

tambourine with a smaller membrane.

TOMS

Toms can be synthesized in one of the two ways depending on how deep and

wide you want the tom drum to appear. Typically, they’re synthesized by using

the same methods as producing a kick drum but utilize a higher pitch with a

longer decay on the amp’s EG and some white noise mixed in to produce

some ambience. The settings for the white noise oscillator generally remain the

same as the amp’s EG for the sine wave (zero attack, release and sustain with a

medium decay). Alternatively, they can be produced by creating a snare timbre

by mixing a triangle wave with a noise waveform but modulating the noise wave-

form with pitch so that it falls while the triangle wave continues unchanged.

PERCUSSION GUIDELINES

Although we’ve looked at the percussive instruments used throughout all dance

genres, if you decide to become more creative, or simply want to experiment

further with producing percussive hits there are some general guidelines that

you can follow.

The main oscillator usually always consists of a sine or triangle wave with its

pitch modulated by a positive pitch envelope. This creates the initial tone of

the timbre while the second oscillator is used to create either the subsequent

resonances of the skin after it’s been hit or alternatively the initial transient.

For the resonance, white or pink noise is commonly used, while to create the

transient, a square wave is often used. The amp and ﬁ lter envelope of the ﬁ rst

oscillator is nearly always set to a zero attack, zero release and medium decay.

This is so that the sound starts immediately on key press (the drummer’s strike)

while the decay controls how ambient the surrounding room is. If the decay is

set quite long the sound will obviously take longer to decay away, producing

an effect similar to reverb on most instruments. That said, if the decay is set

too long on low-pitched percussive elements such as a kick it may result in a

‘ whooping’ sound rather than a solid hit. If the second oscillator is being used

to create the subsequent resonance to the ﬁ rst oscillator then the amp and ﬁ l-

ter settings are the same as the ﬁ rst oscillator whereas if it’s being used to cre-

ate the transient, the same attack, release and sustain settings are used but the

decay is generally much shorter.

For more creative applications, it’s worthwhile experimenting with the slope

of the amp and ﬁ lter’s EG decay and occasionally attack. Also, by experiment-

ing with frequency modulation and ring modulation it’s possible to create a

host of new drum timbres. For instance, if the second oscillator is producing

a noise waveform, this can be used to modulate the main oscillator to reduce

the overall tone of the sound. What’s more, by using the ﬁ lters the sound can

Programming Theory

CHAPTER 4

101

be shaped to ﬁ t into the current loop. For example, using a high-pass ﬁ lter you

can remove the ‘boom’ from a kick drum which, as a consequence, produces a

tighter and more punchy kick. The key is to experiment.

The data CD contains audio examples of these timbres being programmed.

PROGRAMMING BASS

Synthesizer basses are a difﬁ cult instrument to encapsulate in terms of program-

ming as we have no real expectations of how they should sound and they always

sound different when placed into a mix anyway. As such, there are no deﬁ nitive

ways to construct a bass as pretty much anything goes provided that it ﬁ ts with

the music. Of course, as always, there are some guidelines that apply to all basses

and many genres also tend to use a similar bass timbre, so here we’ll concentrate

on how to construct these along with some of the basic guidelines.

Generally speaking, most synthesizer bass sounds are quite simple in design

as their main function is to supply some underpinning because it’s the lead/

vocals that provide the main focal point. As a result, they are not particularly

complex to programme and you can make some astounding bass timbres using

just one or two oscillators. Indeed, the big secret to producing great basses is

not from the oscillators but from the ﬁ lters and a thoughtful implementation

of modulation to create some movement.

Whenever approaching a bass sound it’s wise to have the bass riff programmed

and playing back in your sequencer along with the kick drum and any prece-

dence instruments. By doing so, it’s much easier to hear if the bass is inter-

fering with the kick and other priority instruments while you manipulate the

parameters. For example, if the bass sound seems to disappear into the track

or has no deﬁ nite starting point, making it appear ‘woolly ’, then you’ll need to

work with both the amp and the ﬁ lter envelopes to provide a more prominent

attack. Typically, the ampliﬁ er’s attack should be set to its shortest time so that

the note starts immediately on key press and the decay should be set so that

it acts as a release setting (sustain and release are rarely used in bass timbres).

This is also common with the ﬁ lter’s envelope. Setting the ﬁ lter’s attack stage

too long will result in the ﬁ lter slowly fading in over the length of the note,

which can destroy the attack of the bass.

Notably, bass timbres also tend to have fairly complex attack, so if it needs a

more prominent attack it’s prudent to sometimes layer an initial pluck over the

transient. This pluck can be created by synthesis in the same manner as creat-

ing a pluck for a drum’s kick, but more commonly, percussion sounds such as

cowbells and wood blocks pitched down are used to add to the transient. If

this latter approach is used then it’s advisable to reduce the length of the drum

PART 1

Technology and Theory

102

sample to less than 30 ms. This is because, as we touched upon when discussing

Granular synthesis, we ﬁ nd it difﬁ cult to perceive individual sounds if they are

less than 30 ms in length. This can usually be accomplished by reducing the

amp’s decay (remember that drum samples have no sustain or release) with

the beneﬁ t that you can use amplitude envelopes to fade out the attack tran-

sient oscillator as you fade in the main body of the bass, helping to keep the

two sounds from getting in each other’s way. If this is not possible then simply

reducing the volume of the drum timbre until it merges with the bass timbre

may provide sufﬁ cient results.

Another important aspect of creating a good bass is sonic movement. No mat-

ter how energetic the bass riff may be in MIDI, if it’s playing a simple tone

with absolutely no movement our ears can get bored very quickly and we tend

to ‘turn off ’ from the music. In fact, this lack of movement is one of the main

reasons why some grooves just don’t seem to groove at all. If it’s a boring tim-

bre, the groove will appear just as monotonous. This movement can be imple-

mented in a number of ways. Firstly, as touched upon in the genre chapters,

programming CC messages or using velocity commands can breathe life into

a bass provided, of course, that you programme the synthesizer to accept these

controllers. Secondly, it’s often worthwhile assigning the modulation wheel to

control the ﬁ lter cut-off, resonance, LFO rate or pitch. This way, after program-

ming a timbre you can move the wheel to introduce further sonic movement

and record these as CC data into the sequencer. And ﬁ nally, you can employ

LFO modulation to the timbre itself to introduce pitch or ﬁ lter movement.

If, on the other hand, you decide to use a real bass guitar then unless you’re

already experienced it isn’t recommended attempting to programme the tim-

bre in a synthesizer or record one live. Unlike synthetic timbres we know

how a real bass guitar should sound. If these are constructed in a synthesizer

they often sound too synthetic while recording a bass guitar reasonably well

requires plenty of experience and the right equipment. Thus, if you feel that the

track would beneﬁ t from a real bass it is much easier to invest in a sample CD

or alternatively Spectrasonics Trilogy – a VST instrument that contains multi-

samples of a huge number of basses, real and synthetic.

DEEP HEAVY BASS

The heavy bass is typical of drum ‘n’ bass tracks (by artists such as Photek) and

is the simplest bass timbre to produce as it’s essentially a kick drum timbre

with the ampliﬁ er’s decay and release parameter lengthened. This means that

you’ll need to use a single oscillator set to a sine wave with its pitch positively

modulated by an attack decay envelope. On top of this it may also be worth

synthesizing a short stab to place over the transient of the sine wave. As with

the kick, the best way to accomplish this is to use a square wave pitched down

and use a very fast ampliﬁ er attack and decay with no release or sustain. Once

this initial timbre is laid down, you can increase the sine wave’s amp EG decay

and sustain until you have the sound you require.

Programming Theory

CHAPTER 4

103

SUB-BASS

Following on from the large bass, another alternative for drum ‘n’ bass is the

earth shaking, speaker melting sub-bass. Essentially, these are formed from a sin-

gle sine wave with perhaps a small ‘clunk’ positioned at the transient to help it

pull through a mix. The best results come from a self-oscillating ﬁ lter using any

oscillator, but if the ﬁ lter will not resonate, a sine wave from any synth should

provide good results. Obviously, the ampliﬁ er’s attack stage should be set at zero

so that the sound begins the moment the key is depressed, but the decay setting

to use will depend entirely on the sound you require and the current bass motif

(a good starting point is to use a fairly short decay, with no sustain or release).

If the sine wave has been produced by an oscillator then the ﬁ lter cut-off will

have no effect as there are no harmonics to remove, but if it’s been created with a

self-oscillating ﬁ lter, reducing the ﬁ lter’s cut-off will remove the high-end artefacts

that may be present. Also it’s prudent to set the ﬁ lter’s key follow, if available, to

positive so that the further up the keyboard you play the more it will open. This

will help to add some movement to the sound. If a click is required at the begin-

ning of the note then, as with the drums, a square wave with a fast ampliﬁ er

attack and decay and a high cut-off setting can be dropped onto the transient

of the sine wave. Alternatively setting the ﬁ lter’s envelope to a zero attack, sus-

tain and release along with a very quick decay and increasing the amount of the

ﬁ lter’s EG until you have the transient can also provide great results.

This will produce the basic ‘preset’ tone typical of a deep sub-bass but is of

course open to tweaking the initial click with ﬁ lter cut-off and resonance along

with modulating the sine wave to add some movement. For example, by modu-

lating the sine wave’s pitch by 2 cents with an EG set to a slow attack, medium

decay and no sustain or release the note will bend slightly every time it’s played.

If there is no EG available, then a sine wave LFO with a slow rate and set to

restart at key press will produce much the same results. That said you will have

to adjust the rate by ear so that the tone pitches properly. As a variation of this

sliding bass, rather than modulate the pitch, and provided that the sine wave

has been created with a self-oscillating ﬁ lter, it’s worthwhile sliding the ﬁ lter.

This can be accomplished by setting the ﬁ lter envelope to a fast attack with

no sustain or release and a halfway setting on the decay parameter. By then

increasing the positive depth of the envelope to ﬁ lter you can control how

much the bass slides during playback.

On top of this, keep in mind that changing the attack, decay and release of the

amp or/and ﬁ lter EG from linear to convex or concave will also create new vari-

ations. For example, by setting the decay to a concave slope will create a ‘pluck-

ing’ timbre while setting it to convex will produce one that’s more rounded.

Similarly, small amounts of controlled distortion or very light ﬂ anging can also

add movement. A more creative approach, though, is to use a vocal intona-

tion programme with experimental settings. The more extreme these settings

are, the more the pitch will be adjusted, but with some experimentation it can

introduce interesting ﬂ uctuations.