Synthesisers come in a wide range of shapes and sizes, as physical hardware or as software. The architecture of the synth engine defines how a synthesiser interacts in a creative context, and there exist many concepts available today. This article explains the different terminologies as well as the benefits and downsides of the technologies to help you if you're going to buy your first synths or wish to increase your synth arsenal.
When you are looking to buy a synthesiser, one of the first things you will see on any product description or specification is how many voices a synthesiser has. This equates to the number of notes that your synthesiser can play at the same time.
Now, to start with, you often see terms such as monophonic, polyphonic and duophonic which are all quite clear:
Understanding the terminology
When you read about synthesiser or their technology or you talk with other synth nerds you will stumble across certain terms repeatedly. Let’s take a quick detour and clarify these terms.
Oscillators and Voices
In synth terminology, an “oscillator” would be like a single singer in a choir and a “voice” would be that section in the choir; such as a tenor or soprano section. The more singers in the section, the richer the sound.
If you strip the choir down to one singer per section, you will fundamentally get a different sound than if you have 8 singers in each section.
Voice stealing occurs when your synth runs out of voices (you are playing too many notes at a time) so the synthesiser has to decide which note to stop processing. There are different kinds of voice stealing, such as:
- Stopping the ‘oldest’ note (default)
- Ignoring any new note that is being played
- Stopping the ‘highest’ or ‘lowest’ note
In a monophonic synthesiser, the initial signal is created by one or more oscillator which is then processed by the circuitry (analogue) or digital processors (digital); the entire circuit – including amp, envelope, filter etc. – is contributing to the production of one note. If you press more than one note on your synth, then you will usually jump to the newest note.
A true polyphonic synthesiser will have a specified number of voices; this means you can play up to that number of notes before voice stealing (see below) will occur. The critical feature of a true polyphonic synthesiser is that each voice has its own full circuitry or signal processing and doesn’t share common features like filters or envelops across multiple voices. Basically, a polyphonic synthesiser is the equation of many monophonic synths playing the same sound on different notes.
An example of this is ARGON8, which is an 8-voice polyphonic wavetable synthesiser; each voice – which is made up of 4 oscillators (32 in total) – will be processed through the entire range of processing in the synth – including amp, envelope, filter etc.
A paraphonic synth doesn’t operate in the same way as a polyphonic synth such as ARGON8, SKULPT or CRAFTsynth 2.0. Typically, a paraphonic synthesiser has the same single circuitry or processing as a monophonic synthesiser but with pitch control on multiple oscillators; meaning you can play more than one note through the same amp, envelope, filter etc. This is why some experts argue that paraphonic synthesisers are actually a subcategory of monophonic synths.
It isn’t always audibly apparent that a synth is paraphonic when using static filters, it becomes far more obvious when using dynamic filters, effects and processing (such as delays and LFO’s). When you start to play anything that is not in unison, you get a strange articulation; with a singular articulation structure for all notes. This leads to a one-dimensional sound experience and you will be disappointment by the lack of depth and expression a paraphonic synth produces..
If you compare a polyphonic synth to a paraphonic synth, looking at just ‘voices’ can be misleading; as a paraphonic synth may be able to play more separate notes than a polyphonic synth. The real question is how many voices each synth can fully articulate; in this case, a true polyphonic synth will outperform a paraphonic synth every time.
Where does Modal Electronics stand on Paraphonic Synthesisers?
Modal Electronics does not make paraphonic synths – part of the reason our small synths are associated with such big sounds is because we use multiple oscillators to create a mightier, more textured sound. Making a synth such as CRAFTsynth 2.0 fully paraphonic would mean that we dilute the richness of sound.
Modal Electronics does
not make paraphonic synths
On our digital synthesisers, the ‘SPREAD’ function introduces additional oscillators to achieve unison and stacked octaves, intervals and chords. In the first half of the dial, Wave 1 and Wave 2 are split into individual oscillators and spread out over the frequency spectrum to create a fat unison sound. This effect is increased up to the centre ‘spread’ position. Past halfway, the oscillators arrange themselves into chord shapes.
A huge variation of chords can be created with this control in conjunction with OSC2 detune. However, the average level of the sound will be lower when using the ‘SPREAD’ control because the combining oscillators are phasing in and out and not constantly re-enforcing each other as they would if the spread is set to zero.
Using this setting – you will still achieve richer sound than a paraphonic synth as all chords and intervals are played in unison.
We hope that by reading this, you will get an insight into why we prefer real, or true polyphony to create richer sounds.