Easy Synth Programming – Polyphony
I never heard the word ‘polyphony’ until I worked in a music store. By then I had been playing music for about 13 years. How can something so common go unnoticed for so long? And why is it so confusing?
Real quick: learn to program synths by ear by getting the demo of Syntorial here.
Choir for the Choristers
I challenge you to define the word ‘polyphony’. You’ve got three seconds. Time’s up.
Did you try to break it down into the Latin roots? That can be misleading. Do you really know what polyphony means?
What’s the polyphonic value of a guitar? What about a harp? How many polyphony is your voice?
If you think you can buy pizza with three polyphonys then you are dead wrong. Polyphony is one of those things you don’t need to know until you do. Usually that need comes in a technical form.
One reason you might need to know exactly what polyphony means is you’re a composer. Maybe you’re a music teacher. The other reason is that you’re into electronic music.
I’ll guess you can go far in music without knowing about polyphony. Many synths and samplers come default with polyphony on. Perhaps the first patch is a bass sound and you feel intuitively only one note should happen at once.
What if you’ve played a Moog Voyager? That synth only plays one note at a time. What if you put it through a chorus pedal and now it plays in stereo? Is that still one voice? Is a voice a note?
In this article on polyphony, I want to cover a few things. I want to define this nefariously nebulous word. Then I’m going to show you how to polyphonize Primer and Massive.
Let’s start by defining the word ‘polyphony’. Mono means one. -phony means fake. Just kidding! -Phon means sound. Monophonic operation has just one sound. In synth speak, that sound is called a voice.
I didn’t describe polyphony well in the video. Allow me to correct my mistake here. I will use a choir metaphor.
Imagine a church choir. Usually there are many members in one. Think of the total number of choristers as your ‘max number of voices’. One person per voice just like in real life.
Each choir member can sing a different note. The essential meaning of ‘polyphony’ is ‘different notes’. If you have 30 choristers but only three sing then they use just three voices.
If you’re only going to play triads then why would you ever need more than three voices? Well, you might play some of the notes in a different register. Put a major triad ‘on top’ and another root an octave down and you get four total voices.
You can keep going like this for awhile. With a keyboard, you usually only use 10 fingers maximum. That begs the question of why we would ever need 30 or even 64 voices.
Music has a funny trait in that you never really need it. Sure, you might think you do. Given the choice between that and food the answer is clear. Music also likes to make frivolous demands of itself. That’s partially where so many voices come into play.
Let’s shrink our choir down to just four people. We have the bass singing a root. The tenor sings a fifth. Our alto sings a higher root, and finally the soprano sings the third. All is good in the chapel.
Now lets change from a Imaj chord to a IV major. Two notes change in this case. The third goes up a half step and the fifth goes up a whole step. Again, this is not intrinsically bad.
However, what if we want the two chords to play simultaneously. An example of this is letting the Imaj kind of ‘fade into’ the IVmaj. Something like this might happen if you play the two chords in a very reverberant room.
If you only have four voices then that’s impossible. All the voices need to keep sounding. The ones that change pitch can’t sing two notes at once. What is the solution?
The solution is to add more people to the choir. Remember, the maximum number of voices is governed by how many people are in your choir. Give me two more people, synth voices.
With them I can keep ‘singing’ the third and the fifth. The two new singers can sing the fourth and the sixth. While everyone sings we use every available voice.
There is a synth analog to this. Do you remember your ADSR envelopes? Every voice gets it’s own ADSR envelope.
The ADSR envelope can mimic many sounds’ shape including a human voice. Attack is how they start singing. Decay is how long they take to settle into a final volume. Sustain is how loud they sing the note after decaying. Release is how long they take to let the note fade to silence.
Since every voice gets its own amplitude envelope, we can let one fade as another comes in. Once those notes fade out we are free to start another note. This is how polyphony works.
I also have to admit something. For the intro of this video I didn’t really make a good example of polyphony. That effect is really more like ‘unison’.
We will cover unison in a future lesson. Unison is a subset of polyphony. In polyphony, every voice can sing a different pitch. With unison, in contrast, those voices only double each individual pitch.
Like I said, we’ll deep dive into this later. A brief example of unison will use our four voice choir again. Let’s make everyone sing that big major chord. Now lets add four more people and have them sing the same notes. That’s unison.
Unison is cool because no two singers are exactly alike. If they were then the choir would just sound louder. Due to slight variations in pitch, timing, and other variables we get different sonics.
Let’s talk about how to implement polyphony. I will show you how to do it inside of two synths. Those are Native Instruments’ Massive and Primer. Primer is the synth inside of Syntorial.
I will begin by using Primer as our example. Being a training synth, Primer has a pretty simple interface. You can find its polyphony features at the bottom left.
Primer has two relevant modes here. It has a Mono switch and a Poly one. Try playing with both to observe their effect.
Once you’re done playing around select the poly mode again. Experiment with using a different number of voices. With Primer, the number of voices in the drop down is not your maximum number available. Later, when we set unison modes, we will have more voices available to us.
All polyphony is here is how many different pitches we can play. Remember that each note gets its own envelope. You have to wait for the old note to die out to get its voice back. If you play any sooner then your oldest note will get cut short.
Always keep in mind the envelope factor. Ordinarily you might think this is a bad thing. Actually, by using limited polyphony to your advantage, you can leverage more clarity in your patches.
Let’s think of an example of that. In this week’s Patron Hexspawn Exclusive Ninja Training video, I programmed a messy patch. By limiting my total number of voices, I could cut the mud while accentuating the note of my choice.
By now we have had two examples. The first example was how having more voices can help us. The second example was how limiting our voices a bit can help us. Could limiting our voices to just one help us too? Let’s find out!
When you imagine a full choir singing, what do you hear? You probably hear some dark spots, some swirling altos, and maybe a section of stacatto sopranos. Did you think about the bass?
That bass really shouldn’t stand out much. It needs to remain still to support all the movement above it. If it moves then it needs to do so clearly.
For something to be clear then it needs to be focused. The best way to focus a sound is to have just one playing at a time. Even reverb itself can be considered a voice. Remove even the reverb and you have the utmost clarity.
Similarly, a bass which has been doubled too many times can lose its definition. It’s a bit of an orchestral rule that a bass can be tripled but never doubled. Three basses gives you The Wall of Sound but two just gives you heartburn.
My guess is that you aren’t an orchestral composer. Maybe you are. Maybe you have a passing interest in those styles. If so, now you know the secret to thicker basses. In the case that you’re a Pop/Rap/Rock stylist, read on.
During the course of most songs, I expect to hear a very clearly defined bass. The main exception to this is like my orchestral example. Those sections are exceptions to the rule.
Under a verse or chorus, I want to hear the outline of the harmony. I want a cool riff or booming underpinning of harmony. Give me something I can listen for and sing along with.
Since clarity is defined by the removal of obstacles then remove them! All great mixing starts with great composition. Forget about high passing your guitars – mono your bass!
Specifically, I’m not even talking about stereo width. While you usually don’t want to, add chorus to just one bass. Use your synths monophonic setting, or your sampler’s choke groups, to ensure clarity. You don’t want multiple bass notes inadvertently running into one another.
Secondly, let’s look at another plugin synth. In the VSTi Massive things aren’t much different. The main difference is the number and control of voices it affords. This synth boasts 64 voices to Primer’s paltry 24.
You also get more control over your individual voices. We’ll cover ‘legato’ technique and synth modes later. For now just know that Massive has more knobs.
All you do to set your polyphony here is go into the center window. From there click on voicing. They call it ‘polyphon’ but it’s just polyphony. They probably figure no one will notice.
That pretty much covers everything I know about polyphony and how it relates to synths. In a book called Music: A Crash Course, they say we’ve had polyphony since the 12th century. For something that’s been around for so long this can be a hard concept to understand.
I think the main confusion as to how polyphony relates to synthesizers is the concept of ‘voices’. Since you can set a synth to play in ‘unison’ you have to define these two ideas separately. Synths are great at making you do mental gymnastics.
The first problem I had with synths was understanding the Attack setting. I mentioned this in the amp attack article. As long as you can cope with the foreign nomenclature, synths aren’t too hard.
By now we’ve covered that some words go unnoticed. I guess we hardly notice air too. Sometimes the most common things get taken for granted.
You should have a crystal clear understanding of polyphony by now. Hopefully you can relate this to mono and stereo when you listen to music. It’s a simple concept and it’s time to have it memorized cold.
I showed you how to implement polyphony in two synthesizers. Syntorial’s Primer is a training synth so I started with that one. The first time I did Syntorial I used Analog by AAS inside of Live.
Now that I’m sharing what I’ve learned, I’m going through this course again with Massive. This is something I’m sharing with my Patrons. Donate $3 or more a month to get access to these Ninja Trainings.
As you guys might know, I am a Syntorial affiliate. I heard about it through word-of-mouth. Once I started evangelizing its cause people accused me of selling it.
Well, I wasn’t an affiliate then. I guess if you like something then you’re going to talk about it. If you’re going to talk about something you should try to get paid. Consequently, I signed up to help promote the app.
If you have been waffling about learning synths then it’s time to stop. Just download the demo now and open up the first lessons. You’ll be amazed at what you can learn.
I don’t get paid if you download it. They only give me a commission if you buy it. Only buy the darn thing if you like it. You’ve got nothing to lose.
Detail notwithstanding, they tell me that writing longer articles is better for SEO. I find it difficult to write 3000 words about something. If I’ve left anything out then please leave a comment and tell me.
Easy Synth Programming videos are what I put out every week. You can subscribe on YouTube or with my Hexie Dose Newsletter for updates. I rarely send emails but if you get on the list I’ll reward you somehow.
Check out my original music if you haven’t. I’ve been playing since around 1990 when I took violin in school. Strangely, I was originally into the House music of the time but got sidetracked.
It wasn’t until recently I was able to dive back into it. My stepdad got me into Heavy Metal in the early 90s. That’s why my first EP was pretty much Hard Rock. Not only Rock but also some Electronic.
I think producing Rock music on the computer, especially alone, is woefully inefficient. Synths seem so much more at home in a DAW than a guitar. You can expect to hear more electronic stuff from me in the future.
Anyway, thanks for reading. I hope this was helpful. Like I said, let me know if you have any questions. Most of all I hope you just get started programming synths.
Finally, I publish new material on Mondays 4PM PST. That’s usually an article here on hexspa.com and a video on YouTube. Meet me back here next week for more.