Alesis Fusion Operation instructions

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
The basics....
Since the early days of analog synthesisers back in the 60s, an analog synthesiser can be
broken down into just a handful of basic components. These are:
• SOUND GENERATORS
• SOUND PROCESSORS / MODIFIERS
• CONTROLLERS
The sound generators take the form of OSCILLATORS and also NOISE GENERATORS
The sound processors / modifiers take the form of FILTERS and AMPLIFIERS and also RING
MODULATORS and, these days, effects units such as REVERB, DELAY, CHORUS, etc..
The controllers take the form of ENVELOPE GENERATORS, LFOs (LOW FREQUENCY
OSCILLATORS) plus ‘real-time’ controllers such as KEYBOARD, PITCH BEND and MOD
WHEELS, VELOCITY, AFTERTOUCH, etc..
Don’t worry about the jargon and terminology for now - we will look at these in detail throughout
the course of this tutorial.
The history....
In the early days of synthesisers, all the different components mentioned above were available
as separate ‘modules’ and were connected together using ‘patch cords’:
Thus, to make a sound, you would typically patch a cord (cable) from the output of an oscillator
to the input of a filter. You’d then patch a cord from the output of that filter to the input of an
amplifier and the whole lot would be controlled by various controllers (all patched in with various
cords). Apart from being big, bulky and expensive (not to mention somewhat temperamental
and unreliable!), this made them unsuitable for use live on stage because each sound had to
be made from scratch (there were no patch memories in those days!).
Introduction Page 1

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
With that in mind, in 1971, Dr Robert Moog (largely regarded as the father of modern synthesis)
released the seminal MiniMoog synthesiser. The oscillators, filters, amplifiers and controllers
were ‘pre-patched’ and all connections switched on and off using front panel switches.
Although it had no patch memories to store sounds, the MiniMoog was portable and
considerably easier to use both on stage and in the studio. The MiniMoog spawned all sorts of
similar products from competing manufacturers such as ARP (their Odyssey) and, of course,
Japanese manufacturers such as Roland, Yamaha and Korg who were making their first
synthesisers in the early to mid-70s.
Of course, at this time, all synthesisers were ‘monophonic’
1
- that is, you could only play one
note at a time but in the mid-70s, we saw the release of ‘polyphonic’ synthesisers that could
play as many as eight (!!) notes simultaneously. The first of these was the Yamaha CS80 but it
was Sequential Circuits ‘Prophet 5’ that set the pace for the next generation of synthesisers.
Featuring 5-note polyphony, the Prophet 5 had one more trick up its sleeve - the settings of
the front panel controls could be stored and recalled with a button press allowing you to flick
between different sounds quickly and easily. However, worthy of note is the fact that the
structure of each of the Prophet’s voices was pretty much exactly the same as the MiniMoog’s.
Other manufacturers released similar products all using the same basic layout as the Minimoog
and development stayed rather dormant with no major innovations until 1982 when Yamaha
released the legendary DX7 FM synthesiser. This was also the first synthesiser to feature the
new ‘Musical Instrument Digital Interface’ otherwise known as MIDI. Analogue synthesisers fell
from grace almost overnight and you could barely give them away during the 80s - no-one
wanted that analog sound anymore; instead, they wanted the fashionable FM sounds that
littered almost every record of the time. The DX7 was an immediate success offering 16-voice
polyphony (unheard of in those days), a velocity sensitive keyboard (also very rare back then)
and loads of playable presets for about a fifth of the cost of an analog polysynth!!!
In the mid-80s, affordable sampling also took off with products such as the Akai S900 and so
you’d think the fate of analog synthesisers was sealed.
However, in the 90s, impoverished musicians were picking up these analog relics dirt cheap in
second-hand shops or classifieds and they quickly became popular again. People also re -
discovered their rich, warm and vibrant sound (especially after a decade of clean and detailed
FM and static samples). As a result, prices soon started to escalate and instruments you
couldn’t give away a few years earlier were selling for more than their original price!
However, these old things were unreliable (some notoriously so) and costly to maintain. It was
also difficult to locate good examples of the old instruments (and, of course, they didn’t have
MIDI which by now, had become a vital part of the music making process). But still they
remained popular and these ‘old’ synths were now the ‘new’ things to have!
Advances in computing power meant that is was possible to ‘model’ (i.e. re-create) the sound
of analog synths using advanced digital sound processing (DSP) and so the ‘virtual analog’
synth was born. Offering the characteristic richness of genuine analog synths, they overcame
all of the reliability problems as well as offering greater facilities and higher polyphony at much
lower cost.
Which brings us to the present day. The VA (virtual analog) synth in Fusion is a powerful
engine that offers a truly comprehensive specification that outperforms almost all of the old
analog synths of yesteryear whilst retaining the warmth and character of those old classics.
Let’s now look at the various components that make up a typical analog synthesiser.
1 Some offered the ability to play two notes at a time but were compromised.
Introduction Page 2

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Voice architecture....
The typical signal flow of a typical analog synth was pretty much defined with the MiniMoog and
is something like the following:
Simplified block diagram of a typical analog synthesiser
Two (or more) oscillators generate the basic sound and these are fed into a filter which allows
you to manipulate the tone, often quite dramatically. This is then fed to an amplifier and out to
the audio output(s) on the rear panel. The oscillators’ pitch is typically controlled by the
keyboard but can also be ‘wobbled’ by a low frequency oscillator (for vibrato, for example).
The filter is typically controlled by an envelope generator as is the amplifier and the envelopes
are used to ‘shape’ the sound (i.e. determine whether it is percussive and/or ‘plucky’ or slow
like strings... or just on/off like an organ). Combined, the different permutations of control
settings on even a simple synth allow an astonishing range of sounds to be created.
We’ll look at these different ‘modules’ in turn.
Introduction Page 3

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
SOUND GENERATORS - WAVEFORMS
The oscillators....
The oscillators define the basic pitch and tone of an analog synth sound. You can think of
them pretty much like the ‘strings’ of the instrument. They generate a basic waveform at a pitch
set by a combination of tuning controls, the keyboard and different controllers.
Analogue synths typically offer five different waveforms. These are:
SAWTOOTH A bright sounding waveform suitable for any number of applications such as
strings, brass, pads, leadlines and more.
SQUARE A hollow sounding waveform (not unlike a clarinet).
PULSE A thin and reedy sounding waveform
2
TRIANGLE A mellow sounding waveform (flute-like)
SINE A totally pure sound with no harmonics or overtones - the purest sound known
to man.
The reason these waveforms all sound different is because they each contain different
combinations of ‘harmonics’ or ‘overtones’ which we will look at next. But before that, a brief
explanation of the nature of sound.
All sounds, to a greater or lesser degree, have ‘harmonics’ or ‘overtones’ and it is these
harmonics that define the tone or ‘timbre’ of a sound, the general rule of thumb being the more
harmonics, the brighter the sound (and vice versa ).The predominant pitch we detect in a
musical note is known as ‘the fundamental frequency’ and the harmonics are multiples of the
fundamental’s frequency.
Sawtooth wave
The sawtooth wave is very rich in both odd numbered and even numbered harmonics. That is,
it has harmonics that are twice, three times, four times, five times (and so on) the fundamental
frequency. Thus if the fundamental (and first harmonic) frequency is 500Hz, the 2nd harmonic
is 1kHz, the 3rd is 1.5kHz, the 4th is 2khz, etc.. This creates a very bright sounding basic
waveform and is useful as the basis of many different sounds including strings, brass,
leadlines, basses... in fact, almost anything! The sawtooth shape of the waveform (which gives
it its name) is how the signal would look if seen on an oscilloscope. If you looked at a trumpet’s
waveform on an oscilloscope, it would look similar.
2 On most synths, the ‘width’ or ‘symmetry’ of the pulse wave can be varied for a wider range of
tones. More later.
Sound generators Page 4

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Square wave
The square wave is another very bright waveform but sounds different because it only contains
odd numbered harmonics (X3, X5, X7, etc.). As a result it sounds ‘hollow’ and not unlike a
clarinet. It is useful for many sounds that require that quality and is very useful for reinforcing
bass sounds, especially when tuned an octave down from the other oscillator(s).
Pulse wave
The pulse wave is a bit of an exception because the width of the pulse ‘spike’ can be
continuously varied and the distribution of the harmonics changes according to the width of
the pulse. When the pulse is very thin, the sound is thin and ‘nasal’ (like an oboe) and gets
fuller as the width increases. The pulse wave is good for clavinet-like and other thin sounds.
If you change the pulse width whilst it is sounding, you will hear a pleasing change in tone not
unlike a chorus effect and if you use some controller to do that automatically, this is called
‘Pulse Width Modulation’ (or PWM) and can be useful for creating thick, ensemble textures as
we shall see later.
Triangle wave
The triangle wave is not unlike the square wave in that it only comprises odd numbered
harmonics. However, the harmonics are very much lower in level resulting in a more mellow
sound that is suitable for pure and simple sounds.
Sound generators Page 5

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Sine wave
The sine wave is the simplest waveform known and has only a fundamental with no harmonics
at all. As such, it is a very ‘unnatural’ sounding waveform (there is no sound in nature or musical
instrument that doesn’t contain any harmonics) and therefore is very good for creating pure
sounds. It is also very good for creating ‘sci-fi’ sounds because the early electronic music
pioneers of the 50s and 60s only had very simple sine wave oscillators to play with. The
humble sine wave is also very useful for reinforcing the fundamental of other waveforms and
comes into its own as a ‘sub-bass’ reinforcing the fundamental of a bass sound an octave down
where it’s not so much heard as ‘felt’. This is not a new technique - church organists have been
using it for centuries!!!
Noise generators
So far, we have only looked at pitched waveforms. There are also sounds (such as drums and
sound effects... wind, surf, etc.) that have unpitched elements. These are created on an
analog synthesiser using a noise generator.
Noise is made up of every frequency in the audio spectrum sounding at once. The most
commonly known is white noise, so called because, like white light, it has an even distribution
of frequencies across the spectrum. However, there are also other types of noise suchas pink
noise where the frequencies are balanced across the musical octaves.
The technicalities are largely irrelevant - all you need to know is that white noise is bright and
‘hissy’ and suitable for wind and breath sounds whilst pink noise has more ‘rumble’ and is
useful for thunder and surf sound effects.
Fusion also offers a red noise option which is even more biased towards the low frequencies
and is seriously menacing and ‘rumbly’.
NOTE: Noise has no pitch parameter and is not possible to ‘play’ noise in the conventional
sense of the word - in other words, you can’t pick out a tune using noise!!!
Sound generators Page 6

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Oscillator pitch / tuning / frequency
As well as the basic choice of waveform, there is much you can do to govern the character of
your sound by your choice of oscillator tuning.
A single oscillator on its own can sound a bit sterile, lifeless and ... well ... electronic (although
this can be a quality you might want of a sound). As a result, it is common practice for most
analog synths to have two (or more) oscillators which can be detuned against each other to
give a variety of chorus and ensemble sounds that are ‘fatter’ and ‘warmer’ than just a single
oscillator in much the same way as an orchestral string section has a fuller sound than a single
violin. The amount of detune can be ever so subtle to create a slowly changing sound or can
be quite extreme to give a thick chorus effect.
The diagram below shows the effect of combining two sawtooth waveforms that are very
slightly detuned against each other.
You can see the new combined waveshape constantly changing over time which creates a
much more pleasing and ‘animated’ sound.
As well as small amounts of subtle detune, however, you can tune the oscillators apart by an
octave or maybe two... or you can tune them, say, a fifth (seven semitones) or other intervals
apart.
Of course, the more oscillators you have, the more scope there is for detune and tuning
possibilities. The optimum number of oscillators appears to be three - fewer than that and your
tuning options are limited; more than that can sometimes result in an audio ‘mush’. Fusion’s
VA synth has three oscillators.
NOTE: On Fusion, it is possible to pan each of the three oscillators separately so that as well
as ‘fattening’ a sound with detune, etc., you can also ‘spread’ the sound across the stereo
image with judicious use of oscillator pan.
TIP: Although it is a good idea most of the time to employ two (or more) oscillators to create a
fuller, more animated sound, sometimes a single oscillator is more appropriate. This can be
especially true for creating solid bass sounds where the constantly changing phase
relationship between oscillators can cause the bass sound to lose ‘focus’.
Alternatively, use Oscillator Sync (described on the next page) to lock the oscillators for a solid
sound.
Just those possibilities - combining different waveforms at different tunings - allow you to
create an enormous diversity of sounds.
But there’s more......
Sound generators Page 7

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Pulse Width Modulation (PWM)
We have already seen this mentioned in the description of the pulse wave on Page 6.
The width, symmetry or ‘shape’ of the pulse wave can be varied from an equal square wave to a
very thin pulse
3
as shown below:
When this happens, there is a pleasing ‘chorus’ effect as the harmonic structure shifts and
changes. When setting the pulse width manually, it allows access to an almost unlimited
assortment of different sounds but when put under the control of something like an LFO or an
envelope generator or a real-time controller such as the mod wheel, the possibilities increase
dramatically. Controlled by a cyclic LFO, you can create lush, animated chorus and ensemble
sounds. Controlled by an envelope generator, the pulse width can change over the course of
a note. Controlled by the modwheel, the pulse width can become a performance parameter.
However, unlike most synths, Fusion also provides the same facility with the sawtooth and
triangle waves and instead of a switched choice between one or the other, you can ‘morph’
gradually between the two:
As the waveform symmetry changes between the two extremes, interesting tonal modification
and harmonic movement not commonly available on other synths takes place allowing Fusion
to create some totally unique sounds.
Like the ‘traditional’ PWM described above, this can also be controlled by LFOs, envelopes
and real-time controllers. More on that later.
Oscillator sync
Despite recommendations to detune oscillators to create a ‘fatter’ sound, it can sometimes be
appropriate for the oscillators to be perfectly phase-locked without any detune or ‘beating’. For
example, you might want to set up a solid bass sound with the oscillators tuned an octave
apart. Even if you fine tune them to exactly the same value, there will still be some ‘phasing’
between them - by sync’ing the oscillators, you can achieve the solid sound you want without
the slight detune and potential lack of focus. This is achieved using the SYNC facility.
When this is switched on, the oscillators’ waveform cycles are locked to each other so that they
are perfectly in tune. However, this has some interesting side effects and benefits.....
3In some jornals or articles, you might also see this referred to as the mark/space ratio and it
represents the percentage of time the wavefom is up and down. For example, a mark/space
ratio of 10:90 means that the pulse wave is up for 10% of the cycle and down for 90% of it.
Sound generators Page 8

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
By tuning the oscillators apart in wide intervals with oscillator sync on, many interesting new
waveshapes can be created that can sound distinctly ‘un-analog’ and could almost have been
created on a digital synthesiser. There are no rules for this - simply experiment using different
waveforms and interval tuning combinations.
However, a further side effect and benefit to this is that if the frequency of the sync’d oscillator
is changed during the course of a note, you get a distinctive ‘tearing’ sound not unlike a very
strong flanger effect. You can achieve this manually by controlling the pitch of the sync’d
oscillator using - say - the modwheel and using the effect as a performance parameter (a
popular technique with early synth players such as Jan Hammer or 80s synth pop-meister,
Howard Jones in ‘What Is Love?’) or you can ‘automate’ it using LFOs and/or envelopes to
create many distinctive and classic ‘sync sweep’ sounds.
We will be looking at the use of controllers for this and the PWM sounds later in this tutorial.
Balancing / mixing the oscillators
You don’t always want to have the oscillators at full level all the time - you will want to mix and
balance their relative levels. For example, you might have a sound where one oscillator is an
octave or two up but you might only want a hint of that in the sound or you may want to
emphasise a low octave in a bass sound... or you may have tuned the oscillators a fifth apart
but don’t want the fifth element to be too prominent. Or you may have mixed in a bit of white
noise that needs balancing against the pitched element of other oscillators... whatever.
All synths offer some way of balancing the relative levels of the various oscillators. Some two
oscillator synths have a simple ‘balance’ control (which can be inflexible) whilst other synths
have an oscillator mixer. Some, however, (like Fusion) simply have an output level for each of
the oscillators.
As with most things about an analog synth, there are no rules - just adjust the relative levels
according to taste and the requirements of the sound.
More modern VA synths such as Fusion also offer oscillator pan whereby the oscillators can be
spread across the stereo image for a ‘wider’ sound. This can be particularly useful when
creating certain large, ensemble sounds such as strings and pads to create a wide stereo
sound (but can be a bit overpowering for bass sounds which typically fare better placed mono
and central in the stereo image to create a solid foundation for the track). Again, no rules - just
use your instinct... and experiment!
Sound generators - Conclusion
As you can see, the oscillators alone offer a huge range of sounds to be created even before
we investigate the sound processors / modifiers and controllers. It’s worth getting to know
(and understand) the possibilities offered by the oscillators as they are the building blocks of
any sound... as mentioned, they can be compared to the strings of an instrument and so play
an important part in any sound.
But don’t let this intimidate you - just experiment with different waveform, tuning and mixing
combinations until you arrive at something you like and progress from there. Remember - it
doesn’t matter what you do, you can’t break anything by experimenting!
Sound generators Page 9

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
SOUND PROCESSORS / MODIFIERS
Filters
If the oscillators are the ‘strings’ of an analog synthesiser, the filter is the heart and a synth can
stand or fall on the quality of its filter(s).
Put simply, the filter is one big, drastic tone control that can modify the basic sound generated
by the oscillators. We learned about harmonics in the first section of this tutorial - the filter’s job
is to selectively filter out (or sometimes enhance) these harmonics, thus changing the tone or
‘timbre’ of the sound.
There are many types of filters around that perform different jobs... or rather, have a different
effect on the raw sound they are processing. The most common filters are:
Lowpass filter
Allows low frequency harmonics to pass through unaffected, removing higher frequency
harmonics above the cutoff frequency:
In this example, you can see that all harmonics above the 6th harmonic are ‘cut off’ or filtered
out. This is the most common filter found on ALL analog synthesisers. It closely replicates
nature in that higher frequencies tend to have less energy and so dissipate and die away
quicker than lower frequencies (which is why decaying instruments such as guitar, piano, etc.,
become softer or ‘duller’ as the note dies away). It is also a natural phenomenon that if an
instrument is played (i.e. plucked, bowed, hit, blown - whatever) harder, more high frequency
harmonics are ‘agitated’ and so the sound is brighter (and vice versa - if an instrument is played
more softly, it sounds more ‘muted’). We can use the lowpass filter to mimic these (even if the
sound is overtly ‘synthy’!).
Bandpass filter
Allows a band of harmonics to pass through but removes harmonics either side (below and
above) that band:
From the diagram above, you can see clearly the effect it will have on the sound - the
fundamental is attenuated and harmonics above the 8th are filtered out. As a result, lacking a
strong fundamental frequency, the sound is going to be a bit weak and comprising only middle
frequency harmonics, can sound bright and ‘fizzy’. That’s not to say this filter is not without its
uses - it’s a popular filter in many dance/trance/techno genres for creating bright, ‘fizzy’
leadlines and chordal stabs in anthemic ‘Ibiza’ dance music.
Sound processors / modifiers Page 10

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Highpass filter
Allows high frequency harmonics to pass through but filters out lower frequency harmonics
below the cutoff frequency:
As you can see, the fundamental and second harmonic are filtered out (and the third is
attenuated) which will result in a very thin sound.
Band Stop / Notch filter
This filter type allows lower and higher frequency harmonics to pass through but removes
harmonics in between:
Again, you can see the effect this filtertype will have on the sound - the fundamental and the
first few harmonics are preserved, some upper harmonics are removed but the upper
harmonics remain intact. In practice, the effect of this filter is quite subtle but it can have its
uses and when the notch is moved during the course of a note, it can sound like a mild phase
shifting effect.
Band Boost / EQ filter
This is not so much a ‘filter’ (i.e. a device to remove harmonics) but more of an ‘enhancer’ as it
actually boosts certain harmonics:
In many respects, this filter type can be used almost as a simple tone control for accentuating
certain harmonics in the basic waveform. Its effect is subtle but it does come into its own when
its frequency control is changed during the course of a note.
As mentioned, the most commonly used filter type is the lowpass, probably because
psychologically it corresponds to what we are used to hearing when playing acoustic
instruments - i.e. they tend to get mellower over the course of a note and playing harder or
softer creates a brighter or softer sound respectively. There are other filter types however.....
Sound processors / modifiers Page 11

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Vocal Formant filters
The human voice tract has three frequencies called ‘formants’ that shape the vowel sounds we
make. These formant frequencies move independently of each other and it is their relative
frequencies that create different vowel sounds:
Fusion has three Vocal Formant Filters that boost and attenuate harmonics at certain
frequencies to recreate human vowel sounds. :
When these frequencies remain static, you might hear a certain ‘vocal’ quality but it is when
they move that you can hear something approaching vowel ‘movement’ (for example, moving
from ‘ooo’ to ‘aaaa’ or ‘aaaa’ to ‘eeee’, whatever).
To facilitate this, a controller such an LFO or modwheel needs to be assigned to the filter’s
frequency (which we will look at later) although you can hear the effect when setting these
filters’ FREQUENCY control.
Although these filters won’t allow Fusion to ‘talk’ (!!), they do have a curious, eerie vocal quality
that can be quite endearing in certain circumstances and with certain sounds.
Sound processors / modifiers Page 12

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Modelled filters
Many of the old, original synthesisers’ filters were known for their ‘sound’ - some quality about
them that made them unique or gave them a certain character. The defining element was
invariably the filter. This was often as a result of component intolerances, inadequacies or flaws
in the design and/or other ‘irregularities’. Using advanced digital sound processing (DSP), it is
possible to ‘model’ or recreate the irregularities present in the analog circuitry of these filters
and thus recreate the character of the original.
Fusion currently includes a filter modelled on the one found in one of the classic semi-modular
synths of the early 70s.
Sound processors / modifiers Page 13

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Using filters - Cutoff Frequency
Regardless of their type, all filters pretty much work the same way. There is a CUTOFF
FREQUENCY control that sets the point at which the filter starts attenuating:
When that cutoff frequency is moved by turning the CUTOFF FREQUENCY control, you will
hear the tone of the sound changing. In the example above, when you turn the CUTOFF
FREQUENCY control down, you hear the sound getting gradually softer and less bright as the
upper harmonics are cut. The exact effect depends on the selected filter type but generally,
you hear a ‘wah’ sound as the cutoff frequency changes
Resonance
Another control closely associated with the filter is RESONANCE (also known as EMPHASIS or
‘Q’ on some synths). What this does is boost the area around the cutoff frequency and has the
effect of emphasizing the harmonics at the cutoff frequency:
Note two things - not only is the harmonic at the cutoff frequency emphasised but the
fundamental is attenuated. As you move the cutoff, so each harmonic is picked out individually
especially with increased resonance as shown below:
The resulting sound takes on the characteristic synthy ‘weeeeow’ sound as the cutoff
changes. With certain higher settings of the resonance control, you can actually hear the
harmonics being picked out and individually emphasised.
Sound processors / modifiers Page 14

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Of course, similar things happen with other filter types - for example, a highpass filter:
With a bandpass filter, the ‘band’ becomes narrower with higher resonance settings thus
emphasising the harmonics in the area of the cutoff frequency:
Resonance is an intrinsic component of many analog synthesiser sounds. Unfortunately, in
the early days of synthesis (late 60s, early 70s), it was overused (and employed as a gimmick)
and so the analog synthesiser became synonymous with ‘duck quack’ and ‘strangled cat’
sounds. However, resonance can be used tastefully to create some truly spectacular synth
textures.
Filter slope / roll-off
Another (and final!) aspect to filters is their cutoff or ‘roll-off’ slope.
In the diagrams shown on previous pages, the filter roll-off above the CUTOFF FREQUENCY is
at an angle - it is not abrupt and straight down. This is known as the ROLL-OFF SLOPE and on
modern VA synths, it is possible to define this as an adjustable parameter.
On the original analog synths, the roll-off was usually fixed (although some synths did offer a
switched option for an alternative). The most common filter slopes were:
12dB/Octave Also known as a ‘2-pole’ filter. The roll-off is actually quite gentle and gradually
attenuates/filters harmonics above or below the cutoff frequency
24dB/Octave Also known as a ‘4-pole’ filter, the roll-off is quite steep and attenuates/filters
harmonics above or below the cutoff frequency more dramatically.
Of the two, the 24dB/Octave filter was generally preferred as it has a ‘punchy’ sound and it was
common on many American-made synths from Moog and Sequential Circuits. The
12dB/Octave filter, because of its gentler roll-off, allowed more harmonics above/below the
cutoff point to pass through and so was regarded by many as a bit weak and ‘fizzy’. It was
common on many Japanese-made synths but was also adopted by US manufacturer,
Oberheim (although they offered a switchable 4-pole filter option in later models). Many
Japanese manufacturers followed suit.
Sound processors / modifiers Page 15

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Recent advances in DSP technology means that almost any roll-off can be defined and we can
have anything from 1-pole (6dB/Octave) to 8-pole (48dB/Octave):
And if you are wondering what the dB/Octave refers to..... it’s the amount of attenuation of
level per octave. Thus a 6dB/Octave filter cuts 6dBs (decibels) for every octave and a
24dB/Octave filter cuts 24dBs for every octave and therefore has a more dramatic (some would
say ‘punchy’) effect on the sound. You can largely forget the technicalities though - the rule of
thumb is that a 1-pole (6dB/Octave) filter is going to have a mild effect on the sound whereas a
4-pole (24dB/Octave) or higher filter is going to have a more dramatic effect. In practice, 2-pole
and 4-pole filters are the most commonly used as they are arguably the most ‘musical’.
Filters - Conclusion
To regard the filters as static tone controls is only half the story - they come to life when their
cutoff frequency changes over time. This can be done in several ways but almost always
involves using a controller of some sort such as an envelope generator, LFO or real-time
controller (such as a mod wheel).
Almost all sounds vary in tone/timbre over time and the filter is the ideal tool to mimic that
phenomenon. Even if your intention is not to replicate acoustic sounds, synth sounds can be
significantly improved if they also have harmonic movement and change during the course of a
note - for example, a resonant synth bass sound can benefit greatly from having cutoff
frequency controlled by a decaying envelope as well as having the cutoff frequency controlled
by velocity so that the sound is brighter when played hard and vice versa .
We will come to this soon when we look at envelope generators and later when we examine
‘modulation’. For now, let’s look at another sound modifier.....
Ring Modulator
The ring modulator is a curious device that has been around since the early days of electronic
music. It takes two audio inputs and produces the sum and difference frequencies of those
inputs at the output:
X
Y
RING
MODULATOR
Inputs Output
So, for example, if the frequency of the signal at Input X is 440Hz and the frequency of the
signal at Input Y is 1kHz, the output will have 1.44kHz and 560Hz - the sum and difference of
the two respectively. Mix in the originals and you have a complex sound comprising 440Hz,
560Hz, 1kHz and 1.44kHz.
Sound processors / modifiers Page 16

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Now... because the process is just math, the sum and difference frequencies are inevitably
harmonically unrelated (enharmonic) and thus the ring modulator is very adept at making
discordant sounds.
However, worth bearing in mind is the fact that the ring modulator doesn’t just produce the sum
and difference of the fundamental frequencies of both inputs but also their harmonics. Thus,
feeding harmonically complex waveforms such as sawtooth or square waves into the ring
modulator can produce some truly discordant sounds to the point of being an unmanageable
cacophony! With that in mind, it is often best to just use simple sine or triangle waves with the
ring modulator as you tend to have more control over the discordancy.
Used tastefully, the ring modulator can be used to create beautiful, sonorous bell and chime
sounds. In fact, there is some confusion about the origins of the name ‘ring modulator’ - did it
come from the ring of diodes that was used on the inputs of the original analog designs or the
fact that it produces enharmonic ringing bell sounds?!
The ring modulator can also produce some spectacular early ‘sci-fi’ sounds reminiscent of the
50s and 60s
4
especially if the pitch of one of the inputs changes during the course of a note.
As the pitch of one of the oscillators moves away from the other, you hear a kind of metallic
‘squealing’ sound as the enharmonic sum and difference frequencies change over time.
But the ring modulator has other uses.
As an octave splitter for example. Think about it - feed a 440Hz signal into both inputs and the
sum and difference is 880Hz (one octave up) and 0Hz (no signal).
The ring modulator was (is) also famously used to create the voice of the Daleks from the
classic BBC science fiction series, Dr Who. One input is fed with an audio oscillator generating
somewhere in the region of 30Hz and the other is fed with a microphone with an actor
speaking the lines (typically “EXTERMINATE”!!). The result is a menacing, robotic vocal effect.
Similar techniques can be used using a higher frequency oscillator to create a ‘tingling’ vocal
effect.
And of course, you’re not restricted to ring modulating vocals - try drums or guitar... whatever!
Fusion allows external audio inputs to be ring modulated in this way - simply select EXT IN as
one of the oscillator’s ‘waveforms’ and presto!!!
Granted, the ring modulator is perhaps not the first module to reach for to create ‘mainstream’
sounds for your next foray into the hit parade (although the band Japan skilfully employed
such sounds in their hit ‘Ghosts’) but it is capable of producing a veritable smörgåsbord of weird
and wonderful sounds, especially if very early, pioneering electronica is your bag!
Fusion’s ring modulator goes one step further than ‘traditional’ ring modulators - as well as
being able to ring modulate oscillators 1&2, 2&3 and 1&3, it is actually possible to ring
modulate all three oscillators for some serious sonic mayhem!!!
4 The ring modulator has been around for many decades, actually long before the synthesiser as
we know it and as such, many of the sounds you can create with a ring modullator are highly
reminiscent of and synonymous with early electronic music from such pioneers as Louis and
Bebe Barron who created the first ever all-electronic music film score for the classic movie,
‘Forbidden Planet’, and also the BBC’s Radiophonic Workshop who provided electronic
soundtracks and sound effects for many BBC TV and radio production during the 60s including
Dr Who.
Sound processors / modifiers Page 17

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
Amplifier
Strictly speaking, the amplifier in the audio chain of an analog synth is not really a ‘processor’ as
such other than it allows you to control level, typically at the end of the signal chain before the
signal reaches the outside world.
However, the beauty of the amplifier is that its output level can be controlled by other devices
such as envelope generators and/or LFOs. We will come to this in a moment.
Many (if not most) modern synths’ final amplifiers are now stereo and also offer panning
allowing you to position and/or spread the signal across the synth’s left/right outputs. This may
also be controlled with LFOs, etc., for a wide range of dynamic stereo sounds.
Sound processors / modifiers - Conclusion
So far, we have looked at ways to generate a sound and then modify that in various ways. In
the next phase of our voyage of discovery, we will see how we can change the nature of a
sound over time as we embark upon our first foray into the world of controllers.
Sound processors / modifiers Page 18

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
CONTROLLERS
Before we look at the various controllers, we are going to take a trip back in time to take a basic
lesson in ‘voltage control’ because with an understanding of how synths used to work, we can
better understand how controllers work in a modern environment.
Today, the late Dr Robert Moog is largely regarded as the father of modern synthesisers.
However, synthesisers and electronic music had been around in one form or another for a
some time before Moog brought his products to the market. However, these were often test
laboratory oscillators and graphic equalisers, ring modulators, simple tape delays, etc., and
early electronic music pioneers had to record small snippets of sounds created with this
equipment and, using tape splicing techniques, painstakingly ‘assemble’ a piece of electronic
music. It was a laborious and time-consuming affair as you can imagine and a few seconds of
electronic music could take days to make!!
What Moog did was ‘rationalise’ the process: by splitting the various elements of sound into
different components such as we have so far discussed - sound generators and sound
processors. However, what was unique to Moog’s synthesisers was voltage control which
allowed predictable control of these different components (such as pitch, waveshape, tone,
amplitude, etc.)
5
.
The idea is simple - apply a varying voltage to the control input of an oscillator and the pitch will
change; apply a varying voltage to the cutoff frequency of a filter and the tone will change;
apply a varying voltage to the control input of an amplifier and the amplitude/level will change.
Also, Moog devised the idea of the 1Volt/Octave (1V/8ve) rule - i.e. if the voltage doubles, so
does the pitch of the oscillator (or the frequency of the filter’s cutoff or the amplifier’s output
level):
0V
1V
2V
3V
4V
C0 C1 C2 C3 C4
5 Another synth pioneer, Don Buchla, was also working in similar areas at the time. However, his
designs were maybe a bit more esoteric and aimed more at ‘avante garde’ composers. His
synths also used voltage control but they didn’t always conform to a predictable ‘standard’ like
Moog’s.
Controllers Page 19

ALESIS FUSION ANALOG SYNTHESIS TUTORIAL
With this ‘rule’, it becomes considerably easier to control sounds. For example, if we have a
keyboard that generates 1 volt for every octave, we can ‘play’ the pitch of the oscillators
musically
6
.
And what do you think would happen if we had a device that generated a slowly rising and
falling voltage that was applied to pitch (i.e. oscillator frequency)? Such as this:
0V
+1V
-1V
That’s right - the pitch will slowly rise an octave and then fall two octaves and then rise again,
etc., as the voltage rises and falls. What about this?
0V
+1V
-1V
That’s right - the pitch will rise slowly and then drop abruptly and rise again as the voltage slowly
rises then drops suddenly. And this?
0V
+1V
-1V
Correct! The pitch will jump up and down abruptly. In fact, given the 1V/Octave rule, this would
be an octave jump/trill between the two extremes.
Congratulations! You now understand how the LFO (low frequency oscillator) works!!!
6 It sounds all so rudimentary now but back then, this was cutting edge! Previously, the only way
to pitch an oscillator was to record little snippets of an oscillator (the pitch of which was set
manually) and then splice the bits of tape together to create a ‘melody’!!
Controllers Page 20