ben m
Joined: 15 Sep 2002
Posts: 337
Location: UK |
| Week 4 - Synthesis 1 |
 |
|
Right, this week it is imperative you start writing your own answers rather than cutting and pasting.
As you can see there aren't many questions this week but you are required to answer in some detail as you will need to get a handle on the principles of synthesis.
Good Luck.
Describe each of the following methods of synthesis, and name one famous example/model of
each:
a) Analogue
b) FM Synthesis
c) Physical Modelling
d) Wavetable & S+S
Define the following terms within the synthesis discipline;
a) Oscillators
b) Filters
c) Envelopes and Amplifiers
d) LFOs
e) Velocity Zones
f) ADSR Curves
Explain the following terms with reference to a synthesiser:
a) multitimbral
b) polyphony
c) after touch
d) on velocity
Arrange the following devices in the order they would normally be used for subtractive analogue
synthesis:
a) VCO
b) VCA
c) VCF
FM Synthesis
a) What is an operator?
b) What is a carrier?
c) What is a modulator?
d) What is an algorithm?
How does a digital oscillator produce a waveform?
Briefly outline the basic principles of granular synthesis including definitions of 'grain' and 'cloud'
How would a computer soundcard with onboard sounds usually generate its sounds?
Which types of synthesiser are most suited to which applications?
Synths
Analogue
FM
Sample + Synth
Wavetable
Physical modelling
Applications
1980s synth pop music
Versatile ‘workstation’ keyboard
Emulation of acoustic instruments
1960’s Sci-Fi film soundtrack
Tinkly electric piano sounds
|
Mon Sep 19, 2005 2:42 pm |
|
|
tobyh1000
Joined: 10 Jul 2005
Posts: 13
|
| synthesis 1 |
|
|
Synthesis 1 – Week 4
Describe each of the following methods of synthesis, and name one famous example/model of each:
a) Analogue
Within a synthesizer, analogue synthesis is the method where sound starts out as an analogue signal which is processed by more analogue signal, which then leads to the output port. Basically creating sounds from nothing. The components involved in an analogue synth are: keyboard, Voltage controlled oscillator (VCO), envelope generator, voltage controlled filter (VCF), voltage controlled amplifier (VCA), low frequency oscillator (LFO), modulation wheel and a pitch bend wheel. In essence the 3 main components are the VCO, VCA and VCF.An example of analogue synthesis would be the Roland Juno-60, which was created in 1982.
b) FM Synthesis
Frequency modulation synthesis was discovered by John Chowing of Stanford University in 1973. It is used to make sounds by taking a waveform from one oscillator then modulating that frequency with another oscillator.An example would be the Yamaha DX7, which was created in 1983.
c) Physical Modelling
With this method of synthesis you do not create sound directly like synthesis involving oscillators, but the process that produces the sound instead. You can define the instrument through the process and specific parameters which make the sounds – generally creating a higher level of sound synthesis.An example would be Roland VG-8
d) Wavetable & S+S
Wavetable and S&S (sample and synthesis) sound is not created directly again but instead uses pre recorded samples of real instruments; wavetable has the ability to produce top quality sounds so it follows that the quality of the instruments depend on the quality of the original recordings.An example of wavetable based synthesis would be midi synthesis on a soundcard in a pc.
Define the following terms within the synthesis discipline;
a) Oscillators
Or VCO- voltage controlled oscillator, is the module that produces sound. A constant tone is produced and the pitch is determined by a keyboard input dependant on the voltage value; where a high voltage would receive a high pitch and vice versa.
b) Filters
Or VCF - voltage controlled filter, is the module that controls the tone of the sound by filtering certain frequencies from the sound
c) Envelopes and Amplifiers
An envelope is generated by an envelope generator and basically produces a control voltage that can control the sound shape or sound tone depending on whether it is fed in to the VCA or VCF. The VCA; or voltage controlled amplifier controls the volume of the sound which in itself can also be controlled by an envelope generator as mentioned above.
d) LFOs
Or low frequency oscillator, produces a low output frequency that is used to modulate sounds and is not meant to be heard directly instead being used to create effects such as vibrato.
e) Velocity Zones
Velocity zones basically depict the zone from silence to maximum volume where it is controlled by the keyboard keys-the harder the keys are pressed the higher it will reach in the velocity zone, i.e.; the louder it will be.
f) ADSR Curves
ADSR curves are envelopes that have 4 sections: Attack, Decay, Sustain and Release.Attack is the time the sound will go from silence to maximum volume. Decay is the time taken from peak volume to fade to sustain. Sustain is the volume level of a sound when a key on the keyboard is held down. Release is the time taken from sustain to silence.
Explain the following terms with reference to a synthesiser:
a) multitimbral
Multimbral is when a synthesiser module can play and listen to all the sounds and samples contained on it simultaneously.
b) polyphony
Polyphony is when a synthesiser can play more than one note at a time.
c) after touch
After touch is when a synthesiser module can hold down a key after being played.
d) on velocity
Velocity being the measure of loudness values of notes pressed on a synthesiser module.
Arrange the following devices in the order they would normally be used for subtractive analogue synthesis:
a) VCO
b) VCA
c) VCF
VCO --> VCF --> VCA --> Output
FM Synthesis
a) What is an operator?
An operator is an oscillator, within FM synthesis. Operators are needed to enhance the signal.
b) What is a carrier?
A carrier is the original oscillator that produces the first signal to be modulated.
c) What is a modulator?
A modulator is the oscillator linked up to the carrier so to allow modulation of the carrier signal.
d) What is an algorithm?
An algorithm with regard to FM synthesis is the available combination of routing of signal through the operators.
How does a digital oscillator produce a waveform?
A digital oscillator produces a waveform by recalling electronically stored samples from memory and processing them. Digital instructions are sent dependant on the pitch, sample and effect and then outputted.
Briefly outline the basic principles of granular synthesis including definitions of 'grain' and 'cloud'
Granular synthesis is the method for the synthesizing sounds digitally, where the samples are very short (and usually ‘sound effects’ instead of actual musical sounds). These very short samples (from 1 - 50ms in length) are called grains. Multiple grains layered together are called clouds.
How would a computer soundcard with onboard sounds usually generate its sounds?
A computer soundcard with onboard sounds would usually generate sound by use of FM synthesis. Use of wavetable and physically modeling syntheses are also common but usually with higher spec soundcards.
Which types of synthesiser are most suited to which applications?
Synths with Applications
Analogue
1960’s Sci-Fi film soundtrack
FM
1980s synth pop music
Sample + Synth
Tinkly electric piano sounds
Wavetable Versatile
‘workstation’ keyboard
Physical modelling
Emulation of acoustic instruments
References
http://www.geocities.com/SunsetStrip/Birdland/3547/analog.htm
http://www.harmony-central.com/Synth/Articles/Physical_Modeling/
http://www.geocities.com/SunsetStrip/Underground/2288/2fmsynth.htm
www.wilkipedia.org
www.google.com
|
|
Sat Sep 24, 2005 9:33 am |
|
|
|
|
Polarman
Joined: 24 Jun 2005
Posts: 55
Location: Barbados |
|
|
|
Describe each of the following methods of synthesis, and name one famous example/model of each:
a) Analogue
With analogue synthesis the sound starts as an analogue signal, and is then processed by more analogue signal, before it’s coming out on the output. The first synthesizers were analogue and put together from various test equipment used in laboratories. All the modules were put together into a one single box and called Voltage Controlled Synthesizer. The basic idea is that you connect a keyboard to a voltage controlled oscillator (VCO), a voltage controlled filter (VCF) and a voltage controlled amplifier (VCA). Another word for analogue synthesis is subtractive synthesis. The reason for this is the following:
The VCO generates a tone or sound normally sinus, square wave, saw tooth wave or noise. This wave has different frequency spectrums. To shape the sound you need to use the VCF to take away the unwanted overtones. Finally you need to shape the attack, sustain, sustain and release. One of the most famous models is Minimoog.
http://www.geocities.com/SunsetStrip/Birdland/3547/analog.htm
http://www.soundonsound.com/sos/jul99/articles/synthsecrets.htm
b) FM Synthesis
In one way one could say that FM synthesis the opposite of analogue synthesis. In FM synthesis harmonic overtones are added instead of being removed. The easiest way to explain this is to think of an LFO connected to the amplitude of an oscillator. The amplitude of the output signal varies with the frequency on the LFO. If the LFO is set to 1 HZ the amplitude of the signal out of the oscillator increases the first half second and fades the second half. If you continue to raise the frequency of the LFO the oscillators wave will be distorted. In other words one sinus wave is used to modulate another which results in extra harmonics. One could say that it's a way of mixing two sounds to create a third (the modulation). This technique became very famous with the release of Yamaha DX7. The technology works for both analogue and digital oscillators and it is the same technology used for FM radio.
http://www.computermusic.co.uk/tutorial/fm/fm2.asp
http://www.soundonsound.com/sos/apr00/articles/synthsecrets.htm
http://www.fm-alive.com/fm_explained.htm
c) Physical Modelling
Another word for Physical modelling is Virtual Synthesis. Physical Modelling is different from other types of synthesis. A modelling synthesis algorithm attempts to represent the characteristics of 'real' instruments, using a model from a physical analysis of the sound/instrument. Modelling synthesis can be seen as a 'predictive' sound synthesis method, because the sound is generated refers to the actual physical response of the system. Most other methods of synthesis can be thought of as 'descriptive' sound synthesis, because they use waveform generators and filters in an attempt to emulate the timbre of a real instrument.
http://www.cs.caltech.edu/~ashman/ee107b/project107b.html
http://www.soundonsound.com/sos/jul98/articles/synthschool9.html
d) Wavetable & S+S
This technique use both samples and subtractive synthesis. There are two types of solutions for this.
The synthesizers can use a sample that provides the transient stage of the sound, whilst the sustain stage is synthesized and provided by complex oscillators. The two sound portions are then cross-faded. With the short transient of the sound there is an ability to determine the ‘nature’ or ‘type’ of sound.
The second type use a sample as the sound source (generator), then uses modifiers such as tremolo, vibrato and filters. The original sample may still be ‘recognisable’. There are more complex systems, using a number of samples as sources then separately modifying and mixing them all together, this may result in more sophisticated synthesized sounds.
The samples are normally are stored in a "wavetable" which is the sample memory .The sample memory in these systems contains a large number of sampled sound segments, and can be thought of as a "table" of sound waveforms which may be looked up and utilized when needed. One example of a wave table synth is Kawai K4.
http://www.harmony-central.com/MIDI/Doc/tutorial.html
http://www.acoustics.salford.ac.uk/student_area/bsc1/music_technology_systems/Synth/Lecture_4.doc
Define the following terms within the synthesis discipline;
a) Oscillators
An oscillator in the synthesis discipline is a VCO (Voltage Controlled Oscillator) or a DCO (Digital Controlled Oscillator). This is normally the first step in the synthesis chain and it is here the basic sound is produced. The actual sound in both is analogue. The only difference is that with digital control you get a better stability on the generated tone.
Oscillators can generate one of a number of waveforms like Sawtooth Wave, Square Wave, Noise, Sinus wave The waveform is the 'shape' of the output produced by the oscillator. It determines how the oscillator will sound - in particular how 'bright' it sounds. The brightness is determined by the number and type of harmonics present in the waveform The pitch of this tone is determined by the value of a control voltage sent from the synth's keyboard or LFO - usually the higher the key played, the higher the voltage sent and the higher the pitch. Many synths have more then one oscillator to be able to create a more “fat” sound. .
http://www.sequencer.de/synthaudio/synthesizer-basics.html#resynthese
b) Filters
The filter in a synth is used for shaping the tone from the oscillator (s). The filter does this by removing harmonics. This is a critical part of the synth and some synth are famous because of their filters. The filter (s) are ususally VCF (Voltage Controlled Filter).
LPF - low pass filter - filters out higher frequencies above a certain cut-off frequency with a certain slope.
HPF - high pass filter - filters out low frequencies below the cut-off.
BPF - band pass filter - this is a combination of LPF and HPF: so it cuts upper and lower frequencies, so what’s left is a small "band" in between.
BRF/notch - the opposite of BPF, the band reject filter or so called notch - only filters out a small band, so it is exactly the opposite of a BPF.
APF = all pass filter – not a filter as such only used for phase shifting, mostly used in physical modelling synths.
http://www.sequencer.de/synthaudio/synthesizer-basics.html#resynthese
c) Envelopes and Amplifiers
Envelope is how the sound starts and changes over time in other words the dynamic change. Every instrument and sound has its unique envelope. An organ for example sounds directly when you press a key and stops directly when you release it while a a violin playing pp behaves completely different. The synth uses a Voltage Controlled Amplifier for the envelope and normally you have these settings (ADSR):
A = Attack (time for the sound to reach maximum level after pressing a key)
D = Decay (time to reach sustain level from maximum level)
S = Sustain (level as long as you hold the key)
R = Release (time for the signal to reach silence after releasing the key)
These parameters can also be used to drive the VCFs.
http://www.sequencer.de/synthaudio/synthesizer-basics.html#resynthese
d) LFOs
LFO - low frequency oscillator - this is the same as an oscillator but its not used to create sounds. It can control the pitch of the VCO or the DCO exactly like the keyboard to create vibrato and other effects. Normally it it works with sinus or or triangle wave in the low frequencies from 1 to 20 Hz. The LFO can be controlled by a modulation wheel letting you vary its frequency. It can also control the VCF.
http://www.sequencer.de/synthaudio/synthesizer-basics.html#resynthese
e) Velocity Zones
Velocity zones is used a lot in sampling. The velocity is determined how hard you press down a key on the keyboard. Normally the number range is 0 to 127. Where 0 would mean not pressing down a key and 127 is the maximum registable value possible. It is possible to address different samples to different velocity ranges. So if you press down a key with value 1-10 sample1 is played, 11-20 sample2 is played. Different velocities address different sounds. This is very useful if you want to recreate a natural sounding instrument since an instrument sounds different playing fff from ppp.
f) ADSR Curves
ADSR Curve is the attack, decay, sustain and the release of a signal. An ADSR curve can be set manually or the synth could have a library/presets of ADSR curves.
Explain the following terms with reference to a synthesiser:
a) multitimbral
A synthesizer or sampler is multitimbral if it can produce more than one sound at a time. 16-part multitimbral, means that it can produce 16 different sounds at once. These parts can be assigned to different MIDI channels for independent control.
http://www.sweetwater.com
b) polyphony
Polyphony means a synth or samplers ability to play notes simultaneously. All instruments have a finite number of notes that it can produce at one time. For example, a six string guitar has a maximum of 6-note polyphony, an ukulele 4-notes etc. A synthesizer might be 32-note polyphonic, and so on. The more notes of polyphony an instrument can produce, the more capable it is of playing complex arrangements and chords.
To refer to multitimbral you can say a 16-part multitimbral synth can produce up to 48 notes of polyphony distributed dynamically across those 16 multitimbral parts.
http://www.sweetwater.com
c) after touch
Aftertouch is MIDI data sent when pressure is used to a keyboard after the key has been struck, and while it is being held down or sustained. Aftertouch is often used to control vibrato, volume, and other parameters. The most common is Channel Aftertouch which looks at the keys being held, and transmits only the highest aftertouch value among them. Less common is Polyphonic Aftertouch, which allows each key being held to transmit a separate, independent aftertouch value.
http://www.sweetwater.com
d) on velocity
In MIDI terms, velocity is how fast a key is moved by the player. There are actually two types: "Attack" velocity (normally referred to simply as "velocity") is how fast the key is struck, while release velocity refers to how fast the player lets go of the key at the end of a note. Velocity is often routed to control the volume of a note, how fast of an attack a note might have, or how bright a sound might be.
http://www.sweetwater.com
Arrange the following devices in the order they would normally be used for subtractive analogue synthesis:
a) VCO
b) VCA
c) VCF
Order in a subtractive analogue chain would be: a, c, b
FM Synthesis
a) What is an operator?
Block diagrams in FM synthesis can become very complex. To be able to describe these you normally group a combination of an oscillator plus any associated envelope generators, mixers and VCAs to an operator. An FM synth is normally built up of 4 to 6 operators. An operator is either a carrier or a modulator.
http://www.soundonsound.com/sos/may00/articles/synth.htm
b) What is a carrier?
The carrier is a signal or an operator that’s being modulated by a modulator.
http://www.soundonsound.com/sos/apr00/articles/synthsecrets.htm
c) What is a modulator?
The modulator is an operator that it used to modify the signal of a carrier.
http://www.soundonsound.com/sos/apr00/articles/synthsecrets.htm
d) What is an algorithm?
Algorithms are predetermined combinations of routings of modulators and carriers. There are different algorithms for different sounds.
http://www.samplecraze.com/tutorial.php?xTutorialID=3
How does a digital oscillator produce a waveform?
Its very important to make a difference between a DCO, which been discussed before, and a digital oscillator.
While a DCO creates it sound actual voltage a digital oscillator creats it sound digitally. This can be sampled loops, full samples a single-cycle looped waveform or a waveform created via additive or FM synthesis. (K5000, DX7). The main thing is that the sound is digitally generated, and converted out thru a D/A converter.
http://www.musicplayer.com/cgibin/ultimatebb.cgi?/ubb/get_topic/f/18/t/014399/p/1.html
Briefly outline the basic principles of granular synthesis including definitions of 'grain' and 'cloud'
Granular synthesis is when sounds are broken down into tiny pieces called grains which then are used to form other sounds. A grain would normally be around 10 to 50 ms. If it is shorter its start to sound like clicks. A cloud is a cluster of sound grains. Clouds are significant because they can define an entire texture of grains, whereas the grain is discrete as an individual particle. Even the the grain can be broken down into smaller components like the envelope and the contents. This can be derived from any source. Sine wave, square wave, audio sample, etc.
Closely related syntheses are wavelet synthesis, grainlet synthesis, Glisson Synthesis and pulsar synthesis.
When you discuss granular compositions you could say that a piece is three minutes long and contains 163 clouds.
http://www.granularsynthesis.live.com.au/mthesis/TimOpieM.pdf
How would a computer soundcard with onboard sounds usually generate its sounds?
A computer soundcard with onboard sounds would usually generate its sounds with FM-Synthesis or Wavetable synthesis
http://www.haifa.il.ibm.com/projects/multimedia/audio_video/midi.html#fmsynth
Which types of synthesiser are most suited to which applications?
Analogue - 1960’s Sci-Fi film soundtrack
Most synthesizer were analogue in the 60s.
FM - Tinkly electric piano sounds
Synths like the Yamaha DX7 had this tinkly electric piano sounds.
Sample + Synth – 1980s synth pop music
Most of the popular synthesizer had this synthesis. The sound of the 80s.
http://www.vintagesynth.com/index2.html
Versatile ‘workstation’ keyboard - Wavetable
A workstation needs to have a broad range of “real sounds” and be able to give you a sound that you need quick...you are only composing.
Physical modelling - Emulation of acoustic instruments
Physical modelling is emulation of “real sounds”. |
|
Sun Sep 25, 2005 10:02 am |
|
|
AndyBarber
Joined: 09 Sep 2005
Posts: 18
Location: North Wales |
|
|
|
Posted: Tue Sep 20, 2005 12:42 am Post subject: Week 4 - Synthesis 1
ANDY BARBER
--------------------------------------------------------------------------------
1. Describe each of the following methods of synthesis, and name one famous example/model of
Synthesis is about the creation of sound. A synthesizer is a collection of boxes/modules each with its own simple function. Modules are connected together so sound can be made. They are capable of producing a tremendous range of sounds.
a) Analogue
Method for creating sounds with a synthesizer using analogue circuitry
Uses voltages to represent audio signals with control signals to manipulate them
There are three main stages of an analogue synth:
SOURCES (generator) - modules which can originate an audio signal (oscillators/noise generators)
MODIFIERS (processor) - modules which modify audio signal (filters)
CONTROLLERS (generator) - used to modulate other modules (envelopes/LFOs/oscillators)
A keyboard is used to send a pitch control signal to an oscillator. Analogue synths use voltage controlled oscillators (VCOs) to produce an audio signal at the desired frequency. Next a Voltage Control Filter (VCF) used to control shape of tone of signal. Finally Voltage Controlled Amplifier (VCA) controls shape of output volume
Modular systems - various separate modules (open-ended/unconnected) can be connected together in limitless combinations, modules still based on compact analogue synths
2 areas of analogue synthesis:
Basic concept of analogue synthesizers is called ‘subtractive synthesis’. Manipulation of tone and colour by filtering out (using VCFs) certain frequencies of a waveform created by the oscillators
Removal or attenuation of harmonics from harmonically rich waveforms to create new sounds
Some use additive synthesis where, theoretically, any sound/waveform can be represented by a set of ‘sine’ waves added together. (waveform described it terms of frequencies and amplitudes of its components) Sine waves are the simplest of waves, they are pure and have no harmonics
Analogue synths have basic waveforms that are regular and cyclic. Each has distinct sound determined by their harmonic content. Each wave is based on fundamental frequency (actual pitch we hear) with harmonics/partials which are higher than fundamental (give sound tonal colour).
String on guitar plucked in middle, standing wave created in form of sine wave. Pluck 1/3rd way along a standing wave with 1/3rd the wavelength is created (freq= 3 x fundamental). Standing waves can exist at all integer division of the fundamental (sine) frequency and are called ‘harmonics’.
f = 1st harmonic = fundamental
2f = 2nd harmonic = 1 octave above f
3f = 3rd harmonic = 1 ½ octaves above f
4f= 4th harmonic = 2nd octave above f
Can’t have none integer values as the string is fixed at both ends both at zero amplitude. Principle can be applied to vibrations in a room causing resonances (air can vibrate anywhere except walls, floor and ceiling)
Sine - Pure, only contains fundamental frequency, hollow, woodwind like sound
Triangle - Brighter sounding due to sharp edges
Pulse - variety of tone colours (cycle time is known as pulse width)
Square - Hollow, clarinet like sound
Saw tooth - Bright, rich brassy sound (High harmonic content)
.
Moog’s ‘Modular synth’ was one of the first analogue synths and required manual cabling between modules to create different patches.
b) FM Synthesis (frequency modulation)
FM synthesis is the idea of using an audio signal to modulate another in order to create harmonically enriched waveforms. Simply taking one waveform (the carrier) and modulating it’s frequency with another waveform (the modulator). The carrier can be either another oscillator or a LFO.
A completely new timbre is created. Lots of additional frequencies called side bands are created when modulators frequency is in the audible spectrum. FM theory provides tool to control the number of audible sideband components and their respective amplitudes.. Rich time-variant frequency spectrum is created using far less oscillators and computations than additive synthesis. Sound doesn’t always have a definite pitch (depends on the relationship of the frequencies involved).
MODULATION INDEX
- Output of modulator is offset by a constant, fc, (carriers frequency)
- combined signal used to control frequency of carrier
- if ‘amplitude’ of modulator is zero (d=0) then there is no modulation (carrier output will be fc)
- if modulator amplitude greater than zero the modulation occurs
- frequency of carrier output proportional to ‘amplitude’ of modulator
‘d’ (Hz) - frequency deviation modulation amplitude
fm (Hz) - modulator frequency
fc (Hz) - offset carrier frequency
ac - Carrier amplitude
d and fm drastically change form of the carrier wave
Ratio between d and fm is known as the modulation index ‘i’
i = d /fm
As ‘i’ increases from zero, number of audio partials increases and energy of offset fc is distributed among them
Ring/Cross modulation occurs when the two signals are multiplied together, creating metallic sounds
Yamaha’s DX7 (1983) was first affordable digital FM synth
c) Physical Modelling
Advantage of being able to build synthetic instruments complete with realistic nuances and expressions
Physical modelling attempts to replicate what happens to sound waves inside real-life instruments
An instrument’s physical features are expressed with equations in mathematical terms. Algorithms are used to hypothesis the instruments sound considering all physical constraints and factors.
Most success has been modelling of woodwind (excitation of wave guide) and stringed instruments (excitation of transmission line) both of which use simple wave equations. To build a physical model you need to understand how the actual instrument works
Example of wave guide modelling using a trumpet. Typically there are three main parts that influence the sound, INPUT, MAIN RESONATING BODY, OUTPUT:
1.EXCITATION
Instruments way of creating the initial sound . Input of synth must be able to approximate various ways in which instrument can be sounded by player. Mouthing techniques of trumpet player, blowing softer or louder effects timbre and pitch. Model accounts for constraints placed upon vibrating object (ie. Vibrating lips)
2.WAVE CONTROL
Trumpet player can control pitch by opening and closing valves. Ultimately varying the length of the tube/wave guide. Tube acts as filter and is main resonant body of instrument. Longer tubes amplify lower frequencies, shorter tubes amplify higher frequencies
3.AMPLIFICATION
A trumpets amplifier is the bell which acts as a transformer for the sound. Point at which sound wave reaches most resistance. Air outside has a large acoustic impedance and some of the sound is reflected back into the instrument
Sample based instruments can reproduce realistic sounds but are often static and lack any real performances nuances (subtle/slight degrees of differences)
Physical modelling entered the mainstream in 1993 with Yamaha’s Virtual Modelling VL1 synth
d) Wavetable & S+S (sample + synthesis)
Wavetable covers an enormous breadth of synthesis techniques creating dynamic patches that change over time. It is almost always sample based, taking a number of sounds held in memory combining them together in a multi-layered sequence of samples. This forms a complex tone that drastically changes in timbre over time. Similar to additive synthesis using complex signals instead of sine waves. Can be used to create hybrid instruments. As sound is played different voices fade in and out giving sound its characteristic evolving texture.
Usually four oscillators are used with different samples/voices assigned to each one
All four oscillators are added together to form the wave output
The volume of each oscillator and when it plays can be defined in two ways:
1. ‘wave sequencing’ Cyclic sequence of waveforms blending from one to another.
Waveforms arranged in a linear sequence
Sequence is looped until key released
Volume of each voices pre-determined
Wave sequence on each oscillator doesn’t have to be same length
2. ‘vector synthesis’ Pre-defined path that moves between four points of a square
Each corner represents one oscillator.
The path indicates the relative amplitudes of the four available sounds.
When point moves to one corner only that sound is played,
when in the centre all four sounds play at equal amplitude
Waveforms in memory have their own amplitude envelope curve ensuring smooth transition between sounds with no audio glitches. A good example would be Cameleon 5000, great for ambience and sound-scapes
2. Define the following terms within the synthesis discipline;
Timbres, like an ‘organ’, will sound static if their signal doesn’t change over time. This can be done by filters, amplifiers and effects.
a) Oscillators
Device for generating waveforms. An oscillator in an analogue synth (VCO) is a circuit which creates a single periodic waveform. Generates a steady signal at a preset pitch. Initial pitch is usually variable by octave, semi-tone and fine tuning. Various parameters of an oscillator can be modulated including, pitch, amplitude and pulse width if available.
Oscillators can be used as control sources too, modulating other parameters. The modulating frequency is usually required to be in the sub-audio frequency range and are so call ‘Low Frequency Oscillators’
b) Filters
Synth’s flexibility comes from its filters. As the name suggests filters are used to remove or attenuate selected frequencies from an incoming signal, usually straight from the oscillators. The main variables with a filter is its cut-off frequency, gradient of filter slope (usually 24dB/octave) and resonance (Q)
There are three main types of filter:
Low Pass - Removes all parts of signal above cut-off frequency
High Pass - removes all parts of signal below cut-off frequency
Band Pass - removes band of frequencies around cut-off frequency defined by band-width
Raising the resonance increases any signal near the cut-off frequency
c) Envelopes and Amplifiers
Loudness of the sound can be contoured
Envelopes are used to shape sounds and make them more expressive
An envelope generator produces a signal which rises and falls once
They send varying control signal with a control voltage (CV) to various parameters in the synth (ie. VCA, VCF)
There are controls which set the time the output takes to rise and fall, standard known as ‘ADSR’
Synths usually have at least 2 envelope generators one of which controlling final signal’s amplitude characteristics
When key is pressed a control signal (“Trigger”) is sent to the envelope generators which initiates their programmed envelop cycle. When the key is held down another signal is transmitted (“gate”) to the generators. Gate signal lets envelope generator know the note is still being played and envelope being
generated will continue to be made until gate signal stops. When gate signal stops envelope continues into its final stage usually the release time.
An amplifier is a processor that changes the amplitude of a signal. The amplitude can be set to off (zero amplitude) or to maximum output. Few analogue systems increase signal level provided by that of the oscillators. Amplifiers are basic tool for adding envelopes to sounds
d) LFOs (Low frequency oscillators)
Also called sweep/modulator generators. Basically used as a modulation source for synth’s other function creating periodic effect to sound
Similar to an oscillator but produces signals of a sub-audio frequency roughly 0.1Hz - 20Hz
LFO routed to VCO creates a change in pitch (vibrato)
LFO routed to VCF creates a change in tone and color of sound (trill)
LFO routed to VCA creates a change in amplitude (vibrato)
LFOs also have a wide range of waveforms
e) Velocity Zones
When the speed at which a key is pressed on synth/MIDI controller is measured, it is known to be velocity sensitive. The data collected from the velocity at which the key is pressed can be used as a control signal to determine aspects of the synth’s behaviour. Velocity is measured from 0 - 127 (0 = note off). Velocity zones allow single keys to control multiple actions (ie. Volume of a note, higher velocity = louder signal)
f) ADSR Curves
The basic cycle of an envelope generator can be split into 4 sections, known as ADSR (Attack, Delay, Sustain, Release)
Attack - determines the rate the output reaches its maximum value
Delay - rate the output decays from the maximum to a secondary level
Sustain - sets the value of the secondary intermediate output level
Release - rate at which output falls from sustain level to zero

3. Explain the following terms with reference to a synthesiser:
a) multitimbral
A synth which can play more than one timbre/patch at the same time is known to be multi-timbral
(e.g. An 8 part multi-timbral synth can play 8 independent patches at the same time). Each patch is usually given a separate midi channel or differentiated using velocity/key zones. Multi-timbral analogue synths requires to have more than one sound producing module. With digital synths the modules are virtual with algorithms combining sounds together in real time for output to single D/A circuit
b) polyphony
Polyphony describes the number of independent notes/voices that can played at the same time by a synth
(e.g. Piano is fully polyphonic as all keys can be potentially pressed at the same time)
c) after touch
After touch is the effect of holding down a key and releasing it after it has been played. Synths can measure the amount of after touch as a midi value (0-127) and use it as a control signal.
d) on velocity
Measures how hard a key is pressed on a synth/MIDI keyboard
4. Arrange the following devices in the order they would normally be used for subtractive analogue
1. VCO (voltage controlled oscillators) creates electrical version of audio signal
2. VCF (Voltage Control Filter ) modifies signal
3. VCA (voltage controlled amplifier) amplifies the final signal
5. FM Synthesis (encoding a carrier wave by variation of its frequency)
a) What is an operator?
An operator is a tone generating unit. With digital synths it consists of the input wave, a digital amplifier and an envelope generator. The output from an operator can be used to modulate another operator
b) What is a carrier?
The carrier signal is the source waveform which is to be modulated. With FM radio carrier signal is usually much higher frequency than modulating signal. With synthesis both carrier and modulator signals can be in the audio range
c) What is a modulator?
Modulation - using one waveform to shape or texture a property of another waveform
(process of varying a carrier signal). A device that performs modulation is known as a modulator. A demodulator can demodulate a signal and one that can do both is called a modem
d) What is an algorithm?
Basically it’s a logical step by step problem solving procedure with a finite number of steps
Often they are numerical, examples being mathematical formulas or instructions in a computer program
In digital synths algorithms are used to modify signal data creating new audio sounds
6. How does a digital oscillator produce a waveform?
A digital oscillator is basically a digital version of VCO. A subroutine within a main synthesis program is used to replace a modules circuitry. A digital oscillator is a looped program that generates a stream of computer generated numbers at a steady sample rate. The digital signal produced is a mathematical equivalent of an audio waveform.
7. Briefly outline the basic principles of granular synthesis including definitions of 'grain' and 'cloud'
Granular synthesis proposes that an audio signal is quantised and has a granular texture like the tiles in a mosaic. Based on the theory that human hearing doesn’t have an infinite degree of resolution. With ears and brain only being able to detect pitch, amplitude and temporal changes of a certain quantifiable measurements.
The theory states that all sounds can be made up of many overlapping sonic events (grains) which are barely heard as single sounds, would fuse together to form a composite timbre. The grains are dispersed by frequency, amplitude and their position in the time line. The interaction of the grain parameters define the micro structure of the event/sample. The macro structure (cloud) of the event is generated by the change in the characteristics of its grains
This theory of sound initiated time stretching and has made it possible for digital audio formats to store and reproduce sound as a series of discrete samples
8. How would a computer soundcard with onboard sounds usually generate its sounds?
At the end of the 20th century computers became more affordable and powerful enough to generate digital audio in real time. DSP (digital signal processors) chips can run algorithms capable of creating accurate simulations of physical acoustic sources and electronic sound generators.
Wavetable lookups are used in all signal generation routines. A wavetable is a section of memory that contains a representation of a segment of sound. On each sample period a value is taken from the wavetable and sent to the output where its then processed by the DSP
The onboard sounds on a soundcard are stored as wavetables and are processed in this way to create a digital signal. The digital signal is then converted to audio with an on board D/A converter connected to speakers via an amplifier
9. Which types of synthesiser are most suited to which applications?
1980s synth pop music Analogue, FM
Versatile ‘workstation’ keyboard Sample + Synth
Emulation of acoustic instruments Physical modelling
1960’s Sci-Fi film soundtrack Analogue
Tinkly electric piano sounds FM
http://www.fm-alive.com/fm_explained.htm
http://www.users.globalnet.co.uk/~bunce/control.htm
http://www.thegatherings.org/workshop0604.html
http://www.cim.mcgill.ca/~clark/nordmodularbook/nm_physical.html
http://homepages.kcbbs.gen.nz/gordon/granular.html
http://www.soundonsound.com/sos/1994_articles/jun94/analogue.html
http://www.cs.caltech.edu/~ashman/ee107b/project107b.html
http://arts.ucsc.edu/ems/music/equipment/synthesizers/digital/Digisynths.html
http://www.soundonsound.com/sos/jul99/articles/synthsecrets.htm
http://www.soundonsound.com/sos/jun00/articles/synthsec.htm
http://x2.i-dat.org/~csem/UNESCO/9/
http://industryclick.com/magazinearticle.asp?magazineid=33&releaseid=5638&magazinearticleid=66702&siteid=15
Digital Music Maker issue #5
Computer Sound Design (Eduardo reck Miranda) 2nd Edition |
|
Sun Sep 25, 2005 2:48 pm |
|
|
|
|
|
|
All times are GMT. The time now is Fri May 16, 2008 2:09 pm
|
|
|
|
| |
