Not a fan of one of the genres?

(You can't hide all the genres)

[Theory and basic knowledge] Synthesizers and how they work

For all Hardstyle tutorials regarding DJ'ing and Producing

Postby zanshi » 08 Apr 2011, 05:07

So yea i did this HUGEASS essay about synthesizers for sounddesign class since we also have to do our own reaktor patches etc and blahblah
since i already got a grade on it, i'm gonna leave it to you guys now (:
also as some kind of apology that i've been barely online the last few months cause of crap-webs in my home in NL
have fun reading (AND LEARNING!!!!!!)!

What is a synthesizer and how does it work?

an assignment about the basics of synthesizers and sound generating devices

A synthesizer (or also just called „synth“) is defined as an eletronic device, mostly an instrument, which generates certain sounds via certain parameters and modules. This process is called sound synthesis. It can be either hardware or software.

First experimental synthesizer pre-models got developed in the 1950’ies and were manufactured by RCA. Those „ancient“ synthesizers had so called tuning fork oscillators which created the basic sound which was later being modified to create the desired tone and sound. Another great example for an early electronic music instrument would be the Theremin, which was already invented in 1919 by russian physics professor Lev Termen. This intrument was later on perfectionized by the legendary Robert Moog. Robert Moog was the first manufacturer of synthesizers which were actually affordable for smaller studios and even home use. This synthesizers first got intruduced in 1964 at the Audio Engeneering Society Convention. These monophone synthesizers lead the way to loads and loads of innovation back in the days and soon the first portable synthesizer was born: the Minimoog, which would lead the way from huge modular systems (which often were as big as waredrobes) to compact workstations for on-stage and home use. The first real polyphone synths came out in the 1970’s. But the Yamaha CS-80 and others were pretty complex and very expensive at that time being. In the 80ties synthesizers began to get really popular in popular music and more and more advanced technologies were developed, such as different synthesis techniques and routings. Synthesizers also got cheaper and more affordable. By the 90ties synths entered the digital age and soon nothing seemed to be impossible. Nowadays musicians and sounddsigners can chose from a breathtaking spectrum of different synthesizers and synthesis methods, no matter if hardware or software, digital or analougue (or even both!). Just recently synthsizers crossed the border between VST-Software and semi-analogue, total integration. Modern pop and dance music simply wouldn’t exsist as we know them today if synthesizers were not invented.

Different types of synthesis and synths

Additive synthesis
Show spoiler
Additive synthesis is a method of sound generation based on the Fourier thesis, which says that every tone can be deassembled into sinewaves. It was first introduced at drawbar organs and later on electronic versions of them. A good and recent example for very advanced additive synthesis would be the Camel Audio Alchemy.

Subtractive synthesis
Show spoiler
Subtractive synthesis works in a slightly different manner. Oscillators generate overtone heavy wave forms which are then modified and shaped by filters, envelopes and other modules with the synthesizer or workstatio. Most modern-day synthesizers use this method as it is very intuitive and easy to understand when compared to additive or Frequency Modulated (FM for short) Synthesis.

Show spoiler
FM or frequency modulation synthesis is based on the mathemathical principle of frecquency modulation and mathematical basics of sound. Next to additive and subtractive synthesis, it is considered one of the oldest synthesis methods. FM-Synthesis needs at least 2 oscillators, which are called Operators when used in FM-synthesis, one as carrier and one as modulator. The modulator signal modifies the carrier signal. If this modulation takes place in the human hearing spectrum, so modulations under 20Hz, then a tremolo is audible. If the modulation is going on faster than that then more overtones are audible. If more than 2 operators are used, the term „complex FM-synthesis“ is used. If complex FM is used, then a lot of so called „modulation-matrices“ are possible, the more operartors there are the more complex can those matrixes be. The operators can be routed either as cascade (an operator modulates an operator which modulates the carrier) or in parallel (all operators modulate the carrier), those methods can be combinied too and allow very complex routings. A very famous example for an FM-Synth would be the Yamaha DX7 and it’s software rebuilt by Native Instruments, the FM8.

Wavetable synthesis
Show spoiler
Wavetable synthesis is easy to handle, easy to use and easy to realize. A tone (mostly c’) gets recorded and seperated into attack, sustain, release and decay if needed (for natural sounding instruments) and is bein pitched into all desired pitches. If a very accurate wavetable is needed, more single tones are recorded. The waveforms are now saved into a wavetable and can be used now to, for instance, modulate an FM-signal or get subtracted and so on, but can also be used on their own. A good example for an excellent integrated wavetable would be the Massive by Native Instruments.

Granular synthesis
Show spoiler
This synthesis method is based and used with a certain sampling method. Thousands of so called grains („frames“) are captured and and replayed. This creates the illusion of a single sound, just like a lot of pictures in a certain time create the illusion of a movie. Granular synthesis features the advantage that those grains can be pitch-shifted as well as time-streched. A very exotic granular-based synth is the Malström in Propellerhead’s Reason or Native Instrument’s Absynth, but also Max/MSP can work with granular synthesis aswell.

Linear-arithmethic synthesis
Show spoiler
The linear-arithmetic synthesis was first introduced by Roland in the mid-eighties. This type of synthesis is using the attack sound of a sample to synthesize the sounds basic characteristics. The rest is done by simple addivtive or subtractive synthesis. Back then, this way a lot of realistic sounds could be realized very easily and without taking up a lot of space. One of the best known LA-Synthesizers is the Roland D-50.

Phase-distortion synthesis
Show spoiler
PD-Synthesis was also introduced in the mid-eighties, but developed by Casio. The PD-synthesis was actually intented to be an alternative to Yamaha’s FM-synthesis, but was never able to gain it’s popularity. The synthesis itself is based on the usage of samples with just one single period, which were easy to save at those days. The actual synthesis happened via modulating the read-out time of the captured wave. As prices for sampling-ROM’s dropped, PD-syntheis was no longer necessary that much since Rompler’s had become affordable.

Physical modelling sythesis
Show spoiler
Physical modelling synthesis is one of the most unique, technically complicated and yet most intutive synthesis methods. This type of synthesis is emulating certain instruments or objects via mathemathical parameters and manipulates the sound via more virtually given objects and sound generating methods without breaking the rules of physics. It’s exclusively realised with CPU- or DSP-containing synthesizers such as software or virtual-analogue synthesizers and was commercially first released in 1994 on the Yamaha VL-1. Formant synthesis, the Karplus-Strong-string synthesis and the digital waveguide synthesis are also based on physical modelling synthesis. Two of the best known physical modelling synthesiszers are the Korg Trinity (the OASYS-engine inside of it is physical-modelling based) and my personal favourite synth: Apple Logic’s Sculpture.

Scanned synthesis
Show spoiler
Is a somekind of more advanced and psychoacoustic-based wave-table synthesis. It was developed by Interval Research and first introduced in 1999 and got adapted as VST in the „Scanned Synth VST“ just in 2006, so it can be considered one of the newest synthesis methods. The synthesis method itself is based on a slow dynamic system, those waves can not yet be heard by the human ear. In order to make this audible, the waveshape is scanned periodically and becomes an soudwave. The pitch is determined by the scanning speed, but pitch control and timbre remain seperate. A good implementation of this new and exotic synthesis type can be found in Csound.


Show spoiler
A sampler is a synthesizer wich is based on recorded or synthesized samples. Those samples have to be replayed latency-free. If the replay-time is changed the pitch changes with it. These days most samplers are based on modules and allow to edit samples and modify them via filters, ADSR and are no longer re-playing machines only. A few well known samplers would be Kontakt by NI, a lot of East-West intruments, Halion and the Logic EXS24. Today’s tracker-sequencers like Renoise can be considered samplers too, since a good sampling engine is mandatory for a fluently working tracker.

The basic components of synthesizers

The oscillator is the heart and core of every synthesizer. A synthesizer can have one or more oscillators, always depending on the DSP/CPU-strenght and the desired synthesis method. Generally speaking an oscillator is an elecrotechnical device generating a waveshape. If this waveshape is periodically, then the oscillator periodically repeats the signal until the user tells the device to stop. Some oscillators feature something called „Unison“.
Unison creates copies of the selected (or all) oscillators and influences the sound to make it sound fatter. The parameter „Unison detune“ on most synthesizers with this feature, controls the amount of detune in between the copies and can make the sound even more fat or aggressive.

The different waveforms:
Show spoiler
The sine wave is the most basic waveshape there is. It’s mathemathical funtion is defined as y(t)=sin(omega*t+phi). The sine wave is just one harmonic and does not actually exsist in nature, it can only be generated in a electronic or digital enviroment. The sound of a sine wave can be compared to the sound of a tuning fork. The sine wave is also 'absolutely essential' when creating a nice subbass to exsisting sounds, since it is able to create and add a bassy sound without disturbing the other harmonics of the sound.

The saw wave is the most harmonic waveshape. It consists of all natural harmonics. The saw wave is one of the most popular waveshapes for creating string-like intruments and sweet sounding soundscapes.


The square wave is a very bright sounding waveform. The sqare wave consists of all integer odd harmonics. It’s sound can be described as very harsh, tho most people associate the sound with Gameboys and nostalgia gaming classics.


The triangle waveshape is a bright sounding wave with just odd harmonics. It sound kind of something in between the square wave and the sine, but much softer since the harmonics decrease steeper than in the square wave spectrum.
Most of all nowadays consist also of a noise-generating oscillator. The two most common types of noise are white noise and pink noise (or 1/f-noise). White noise is a very bright sounding type of noise which evenly generates sound on all possible frequency bands, whilist the pink noise sounds much darker and more even to the human ear. Noise is used to create percussive sounds or add „hiss“ to a patch.

Wavetable, Samples & other

Those are saved on the ROM or patch of the synth and can be loaded into the oscillator. This can either be single, periodic waveshapes or entire *.wav-files (or similar), always depending on the type and model of synthsizer.

The filter section, different filters and the equalizer
The filers of a synth are next to the ADSR the most powerfull modules to shape and change the sound (especially in subtractive synthesis) to the desired timbre and movement. Some synthsizers have such unique filters that it makes a major difference on the sound. All filters have a steepness (Q or emphasis) which can mostly be handled and controlled by the user, for example: 6db, 12db, 18db, 24db and 48db per octave, which are the most common. The resonance of a filter sets an accent to the edge of where the steepness begins. A very popular expamle for very unique filter section would be the legendary TB-303 by Roland.
Show spoiler
High- & low-pass
The low and highpass are the most basic filters. The lowpass filter filters away high frequencies and leaves the sound very bassy and fat, whilist the highpass filter filters away the low frequencies. The sound of a lowpass filtered sound sounds like the sound was played trough a wall or a thick piece of fabric. A highpass filtered sound sounds very bright and sounds like the sound was played trough a telephone.


The bandpass filter filters away high and low frequencies at the same time and leaves the sound with some kind of „radio“ effect. The name bandpass is taken from the fact that this kind of filter only lets a certain frequency-band trough its algorithm or ciruit.

The notch filter can be seen as the opposite of the bandpass filter (since it filters away a frequency-band, instead of letting it pass) or some kind of combination between low- and highpass. It filters away a „hole“ in between two frequencies and leaves the sound with only the desired low and high end of the entire frequency band.

The comb filter filters a certain group of signals. This mostly happens by creating a just slightly time-delayed copy of the original signal and mixing it in. The name comes from the appearance in the spectrum, making it look a bit like a comb.

The equalizer
The equalizer (or EQ) is basically a algorithmic device to either lower or amplify certain frequency bands. An EQ can be graphic or parametric. A graphic equalizer has a „knob“ for every changeable frequency band allowing very precise adding and removing of harmonics and frequencies. A parametric EQ is more flexible. It allows the user to change the steepness and range of the created or allowed frequency bands. Most EQs in synthesizers nowerdays are parametric since they allow much more creative sounddesign. It should also be mentioned that EQs in general are very versaitile and important devices in audio in general.

The low frequency oscillator (LFO)
The LFO or low frequency oscillator is a moving parameter which low frequence (below 20Hz) which can be used to automatize certain „events“ in the synth. Such as automatizing a filter to generate a movement in the sound which is generated by the synthesizer. LFO’s can have waveshapes too such as sines, which is most commonly used and integrated in almost all of nowerdays synths. A good example of the usage of LFO’s are filter automatisations via LFO’s in Dubstep and Drum’N’Bass basslines to create the famous „wobble sound“.

Sequencer and appreggiator
The arpeggiator or simply „Arp“ is a very common feature on synthesizers. It moves a certain note up and down in a selected time and musical space. This can either be a chord or a note. It’s parameters feature the functions „up“, „down“ and „up & down“ which determine the direction of the desired arpeggio. Furthermore there can be a number of octaves selected which the arpeggio will take place in. So, for instance, an arpeggio whith the selected parameters „up & down“ and „2 octaves“ will move the note or chord up and down in two octaves. A sequencer or also called step-sequencer (for drummachines) is a more complex feature. It is based on the features of the arp, but it allowes the user to select the notes and rythm of the triggered arpeggio very precisely. This allows the user to let a melody or chord sequence be played by the synthesizer.

The envelope or ADSR
The envelope is one of the most noticable and powerful elements to shape the sound. It determines the volume fading times and output levels of the sound created. ADSR is the ancronym for the most commonly changed parameters in the envelope. A stands for attack, it defines the time that the sounds needs from triggering it to it’s peakpoint. D is for decay, the timespan the sound needs to fall to the set sustain level. The sustain level is referred to as S and defines the regular output level when the key is held. The release time or R, is the timespan the sound needs to fade out completely when the key or MIDI-note is released. A very common variation of the ADSR is the AHDSR, which also features a „hold“-parameter. Iit allowed to hold the peak of the attack for the in the parameter defined time. Also a very new, but common parameter is the „delay“ which would delay the time before the attack hits in. This way up to 8 parameters can be set in the envelope, but also „degenerated“ envelopes were and are common, for example in the legendary TB-303 which just features a AD-envelope. The functions of an envelope can be as different as the envelope itself. Envelopes can be assigned to every parameter which allows modulation, always depending on the synthesizer though. For example the mix of the filter, the pitch or most commonly the output level of the amplifier. Playing with the parameters of the envelopes can turn a plucky sound into a smooth pad, a screaming stab into a soft sounding musical instrument or a simple sine into a kickdrum.

From the time on where sound-generating-only synthesizers became all-in-one workstations for siundesigners and musicians alike, effects were not allowed to be missing in synthesizers. I’ll quickly explain the most common ones for nowerdays synthesizers, focussing on software and virtual-analogue hardware.
Show spoiler
Chorus / Flanger / Phaser
These three effects are very common and belong to the phase shifting effects. The phaser mixes two audio signals together and shift the phases of one of the signals. The flanger does the same but with a more comb-ish manner since the signal is not phase-shifted but a little bit delayed, which creates a comb in the spectrum. The chorus combines both elements.

Delay / Reverb

A delay or echo device stores a certain amount of signal and plays it back again, creating a echoish soud. A reverb, reverbaration or hall effect uses similar algorithmics to create a single signal refelecting sound.

Distortion comes in many, many, many different froms. From simple overdriving to complex multiband algorithmics everything is possible. The main purpose of distorion is to distort the signal and add a certain timbre or crunch to it.

Tremolo / Vibrato
Tremolo and Vibrato are very classic effects, mostly controlled by an LFO. They either slowly change the amplitude (Tremolo) or pitch (Vibrato) of the sound slightly, depening on how the parameters were set.

The amplifier
The amplifier (or also simply called „amp“) regulates and amplifies the generated and processed signal. It is mostly referred to as „master output“ nowerdays. My default the amp is set to ±0db, but it can mostly be turned from -∞dB (silent) to +6db.

Generally speaking, synthesizers are electronic or digital sound generators which use certain digital algorithmns and/or electrotechnic routings to generate sounds with different methods which we call „synthesis methods“. I would have been able to write even more about the certain synthesis methods and their special capabilities, but this would have been more subject of advanced synthesis and not the basics of synthesizers. Furthermore there is a lot more to tell about special synthesizers and their build-ups, such as workstations and modular systems and the real differences in between analogue and digital, which would have been ore a subject of advanced history of sound synthesis and synthesizers. Anyways, I hope this essay was a nice read. (:
"ars ex scientia"
zanshi (28)
State Hero
Posts: 3369
Location: Netherlands (nl)

Postby Lop-Sided » 09 Apr 2011, 22:38

Man, this must've taken a long time to put together and write!
And it's full of great information, even if it is only the basics! :)

Thanks a lot for posting this! _O_
Great job! :D
Trance producer/Radio Show Host at 90.3FM KDVS
Check out my tuneessssss, broooooo!!
User avatar
State Celebrity
Posts: 1503
Location: United States (us) <-- points to flag

Postby RedRaven » 11 Apr 2011, 03:51

Nice, mad props on putting this together and posting it!
State Newcomer
Posts: 10
Location: United States (us)

Postby MKN » 11 Apr 2011, 15:39

Sick post! :D Well done, must have taken you a while, hope people appreciate it ^^
User avatar
MKN (26)
Posts: 3437
Location: Wales (wa) Llanelli

Postby Shadow Interaction » 12 Apr 2011, 14:35

_O_ _O_ _O_
User avatar
Shadow Interaction (32)
Posts: 5668
Location: Netherlands (nl) Rotterdam

Postby Noxit » 12 Apr 2011, 22:19

:dance: :dance: :dance: :dance: :dance: _O_ :dance: :dance: :dance: :dance: :dance: ||
Open for collabs, PM me with decent work! (Cubase5)
User avatar
Posts: 277
Location: Netherlands (nl)

Postby Aztecz » 13 Apr 2011, 16:11

Awesome post! I'm printing this out so I can do something productive in math class :p
Aztecz (26)
State Tourist
Posts: 72
Location: United States (us)

Postby BrannMannSam » 27 Apr 2011, 17:23

Aztecz:Awesome post! I'm printing this out so I can do something productive in math class :p


Just read the first few segments and this is really helpful :) It is WAY batter than any "professional" guide i have seen, i would love for you to write about more topic if you feel for it :) Gj
State Outsider
Posts: 3
Location: Norway (no)

Postby zanshi » 28 Apr 2011, 00:15

expect me to write some more :D
"ars ex scientia"
zanshi (28)
State Hero
Posts: 3369
Location: Netherlands (nl)

Postby zanshi » 21 Aug 2012, 17:11

uploaded the EXTENDED german version again here:
"ars ex scientia"
zanshi (28)
State Hero
Posts: 3369
Location: Netherlands (nl)

Quick Reply

Register or login to share your passion for the harder styles

  • Related topics
    Last post

Who is online

Users browsing this forum: No registered users and 10 guests