Wiki: Zircon's ReMixing Tips Compendium/Part 3: Synthesizer Basics

Revision as of 22:45, 18 April 2007 by Liontamer (talk | contribs) (link change)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Move back to Zircon's ReMixing Tips Compendium, Part 2 - Working with "Real" Instruments
Continue onto Zircon's ReMixing Tips Compendium, Part 4 - Effects Processing...


Covers the basics of subtractive synthesis, the important controls, and practical applications.

Overview

When someone asks you for critique, one of the most difficult things to address is the use of synthesizers. There are so many methods of getting synthesized sounds, and so many actual synths, that it's next to impossible to give any sort of specific advice on the topic without knowing exactly what the artist is using. To compound the problem, it's hard to know whether they just used sampled loops or notes, or actually created a sound from scratch by themselves. I'd like to try to tackle this problem by going into the basics of creating synthesized sounds by using a free (but high quality) virtual synth that can be used in almost any sequencer with "VST hosting" capabilities.

To start with, I think it's important to briefly describe the type of synthesis we'll be dealing with - subtractive. This method is VERY widely used because it is highly flexible and relatively easy to learn. Though I'll be doing demonstrations on the free plugin "Synth1" (link at the end of the post), the same techniques apply to basically ANY other subtractive synthesizer, and there are lots of them. Anyway, subtractive synthesis basically has one or more basic waveforms being generated by an 'oscillator'. Think NES or Gameboy sounds when you try to picture these basic waveforms (sine, square, triangle, saw). The waves are then routed through some sort of a filter, which removes a certain part of the sound. Also important to the process are envelopes, but we'll get to all this later.

The Basic Waveform

So, let's examine the basic waveforms to begin with. When you're designing a synthesizer 'patch', step 1 should be to select the waveforms you want to use.

  • Sine: the most basic sound. It is very 'mellow', sounding something like an organ on the lower notes, and a whistle on the higher notes.
  • Triangle: slightly more complex than the sine wave. The sound is somewhat 'denser' or brighter than the sine, but still fairly subdued.
  • Saw: significantly more gritty and bright than the sine and triangle waves. Unaltered, it has a sort of a buzzing sound to it, and is great for the basis of lead sounds or cutting/aggressive sounds.
  • Square/Pulse: the most rich of all the basic waveforms. It has the qualities of the saw, except with a more accentuated bass presence, and doesn't sound quite so buzzy. The basic square wave sound is one that is extremely common in NES-era music, so you should recognize it quickly. Another thing to note about the square wave is that its timbre (or overall character) can be changed using a 'pulse width' control.

Oscillators

Now, when you combine two different waveforms, you get an entirely new sound, giving you a wide palette of basic sounds. If you have two oscillators both using the save waveform - say, a saw wave - you will get a different sound as well. This is called 'unison' and it is very popular when creating thick, rich sounds such as those found in dance/trance leads or background pads. In addition, each oscillator can be tuned differently; either using a 'Fine' tune control (for very small values, less than a half step), or the 'Coarse' tune control (which goes one half step at a time). Messing with tuning can result in different sounds as well. For instance, making one oscillator play one octave (12 semitones or half steps) below the other will make the sound denser. Taking two saws and setting one ten cents down and one ten cents up using the Fine tuning will create a modified unison effect that is a staple of trance music.

There are a few other controls worth mentioning in the Oscillator section. These controls affect the most basic level of the sound - before it even gets to the filters or envelopes.

  • Sync button: This forces the oscillators to sync to eachother. The exact technical details of this are a bit hard to describe, but suffice it to say that it creates a very interesting sound, particularly when the oscillators are significantly different in pitch.
  • Ring button: This causes on oscillator to modulate the VOLUME of another oscillator, creating a new kind of sound.
  • FM knob: This controls the amount of frequency modulation. When this is above 0, one oscillator will be modulating the frequency of the other oscillator. If you try this with basic sine waves, you will notice the sound is very similar to the old Genesis music, or even old PC games - and you'd be right, because many old PC games, and particularly the Genesis, had synth sounds created using FM synthesis. FM really introduces a whole new range of sounds, because it introduces properties in the sound that wouldn't have been there before. Experiment with this!
  • Mix knob: The mix between the two oscillators - all left means only osc 1 is playing, all right means only osc 2 is.
  • P/W (Pulse Width): The pulse width of the square waves, if they're being used. This basically changes the characteristic of square waves. Many of the old NES sounds were created by using square waves of different pulse widths.

Modulation Envelope

Finally, the first oscillator can be used to modulate various parameters even FURTHER, using the "M. Env", or modulation envelope. Here are the controls to consider.

  • Destination: Osc1 can be sent to modulate Osc2 (the pitch of oscillator 2), FM, or the pulse width.
  • Amount: How much modulation osc1 is doing, and in what "direction". A value of 0, even if the M. Env button is on, means no effect.
  • A/D = Attack/Decay: These are elements of an envelope. For what these do exactly, you will want to scroll down to the envelope section directly below. After you have read this, you will understand what I mean when I say the modulation envelope has a sustain level of 0 and a release of 0.

The next important element of designing a synth sound is your envelopes. Let's examine the "Amplifier" part of Synth1, also known as the "Amplitude Envelope". An envelope essentially shapes how a sound progresses over time. There are four main elements to the envelope, which are fairly intuitive:

  • Attack (A): This is the amount of time it takes for the sound to reach it's INITIAL volume. An attack of 0 means that the sound will instantly play at max volume as soon as it is played. A longer attack means the sound will fade in. Examples of sounds with short attacks would be basses, plucked sounds, and arpeggio sounds. Sounds with long attacks would mainly be pads or string-type sounds.
  • Decay (D): Once the sounds Attack time is complete, it will then take a certain amount of time to "decay" to another sound level. The Decay knob will control how long it takes to hit the Sustain level, that is specified below. You usually don't have to worry about this value too much, unless you are using sounds with low Sustain levels.
  • Sustain (S): The volume level a sound will stay at if you continue to hold it, and its Decay time is complete. Typically, plucked sounds have a very low sustain, or no sustain at all. Pad sounds, leads, and basses all have relatively high sustain levels.
  • Release (R): Once the sound stops playing (eg. you let up the key on a synth controller), this is how long it takes for the sound to fade out. A Release of 0 means the sound stops instantly. A long release means the sound will slowly fade away, ideal for pad-type sounds.

In addition, you'll see..

  • Velocity control (Vel): This controls how much velocity (how hard a key is struck) affects the volume of the sound. A value of 0 means velocity has no effect. A very high value means that even slight changes in velocity will make a big change in the sound's volume. A middle value (about 64) is usually preferable.

I'll get into some practical applications of envelopes a little later. For now, though, let's move on to another important component of the sound, the filter...

Filters

Filters are essential in sculpting the sound. Basically, a filter REMOVES some element of the sound - this method of synthesizing sounds is referred to as "Subtractive Synthesis". Most synthesizers are subtractive, and even those that aren't (eg. Wavetable, Additive, FM, Granular) usually have filters of sorts. First thing you need to choose when you are editing a filter is the type of filter. Here are the types available:

  • Lowpass (LP): Cuts off all sound above a certain frequency.
  • Highpass (HP): Cuts of all sound below a certain requency.
  • Bandpass (BP): Cuts off all sound above and below a certain frequency range.

You will also see numbers to the right of the filter types; 12 or 24. These basically show how "strong" the filter is, for lack of a better word. When you set the cutoff frequency, or simply "cutoff", the sound does not instantly stop at that point. Rather, the sound gradually tapers off from that point along a slope. A 12 (technically speaking, db/octave) type filter has a gradual slope, while a 24 has a steeper slope. Thus, an LP24 is preferable when you want a sharper cutoff of the sound. Other filter controls that are important:

  • Cutoff (Frq): The frequency at which the sound will be tapered off, as mentioned above.
  • Resonance (Res): For LP/HP filters, how much the volume will be increase at the cutoff frequency. High resonance values can be ear-piercing, so be very careful - however, resonance used tastefully can turn a boring lead into a screaming, awesome one. For the BP filter, resonance controls the cutoff frequency range.
  • Saturation (Sat): A sort of overdrive effect on the filter. Turning this up literally saturates the sound, making it thicker and harsher. When combined with resonance, this can create a heavily distorted sound, similar to the effect of running an electric guitar through an amplifier and distortion or overdrive pedal.
  • Key Tracking (Trk): This controls how the pitch of the note played affects the cutoff frequency. For example, a middle value of 64 means that the note played has no effect. A high key tracking value means that the lower you play on the keyboard, the lower the cutoff scales. A low key tracking does the opposite.
  • Velocity (Vel): This toggle controls whether or not your velocity affects the cutoff. Lower velocities = lower cutoff.

Last but not least, filters have an envelope, just like the Amplitude envelope I described earlier. The only difference is that the filter envelope affects the progression of the filter sound. It's a bit difficult to describe, so you'll have to tweak to hear exactly how it works. However, my examples will demonstrate some filter envelope tricks too.

Low Frequency Oscillators (LFOs)

The final basic component of Synth1 is the LFO section. LFO stands for "low frequency oscillator". LFOs do not actually generate the sound - they simple alter other parameters, such as filter cutoff, or pitch. They're flexible in how much effect they have, and how fast they oscillate. The first component of an LFO is the waveform, which can be selected by clicking on the button to the immediate left of the 'dst' button on Synth1. The most standard waveform is the Sine, which modulates a parameter smoothly. The saw, triangle, and square waves are harsher, while the other two waveforms are more or less completely random. Other LFO controls:

  • Speed/Frequency (Spd): The rate at which the LFOs are oscillating. Self-explanatory.
  • Amount (Amt): How much the LFO actually affects the sound. If the amount is high, an LFO has a very drastic effect - if you want a subtle change in the sound, very low values are better.
  • Tempo Sync, Key Sync buttons: These let you synchronize the cycling of the LFO. Tempo sync means that the LFO will sync to the master tempo of the song - changing the Speed control will then change how often it oscillates per beat (etc). Key sync is the same concept, except the LFO operation changes based on the pitch of the sound created. Tempo sync is often very useful, while key sync not so much.
  • Destination (Dest): For Synth1, there are many different destinations - parameters on the synth that the LFO can modulate. Osc2 is the pitch of oscillator 2. OSC1+2 is the pitch of both oscillators. Filter is the cutoff frequency of the filter. Amp is the overall volume of the sound. P/W is the pulse width of the oscillators' square waves (if they are set to that). FM is the amount of FM between the two oscillators.

LFOs are not essential to the design of synth sounds, but they can be very useful in creating dynamic sounds and those that change over time. I will again be demonstrating examples of using LFOs (for fun and profit).

PHEW! That covers the very basics of this synth (and most subtractive synths), but there are just a few more things I'd like to cover before we get to the actual sound examples...

Play Mode and more

First, let's check out the "Play Mode" section. Take note of the "Mode" button first, which has three LEDs under it.

  • "Poly" means you can play multiple notes at the same time.
  • "Mono" means you can only play one note at the same time.
  • "Legato" is like mono, but a glide effect can be simulated between notes to make the synth play in a smoother manner.
  • The "Portamento" knob will determine how much of a glide time there is between notes.
  • "Unison" means that the synth will be generating multiple notes (in unison) at the same time when you just hit one. This creates a thicker, fatter sound that is slightly more detuned. Very common in trance music.

There are three other sections which I'll describe briefly. You have the Equalizer, which you SHOULD be able to identify - if not, it's simply a way of emphasizing or de-emphasizing parts of the sound. The Tempo Delay creates echoes after you hit a note. The Chorus/Flanger is like the Unison control.

Example #1: Synth Lead

FINALLY! Ok. Now what you really wanted to see (hear?).. the example stuff.

http://www.zirconstudios.com/Synth%20Example%201.mp3

Here's an example of a simple synth lead. What makes it a lead? Well..

  • The amplitude envelope. Short attack, no release, high sustain.
  • The filter. It's lowpass, because we don't want too much of the high stuff to be blasting our ears. Decently high saturation and resonance to make it have some substance.
  • LFO 1 is pointing towards the filter. This makes the sound dynamic over time.
  • Play mode is Legato with some basic portamento action. This makes the sound glide smoothly.
  • Saw + Square waves with "Sync" on. Creates the basic timbre of the sound.
  • A little bit of delay to make things interesting.

http://www.zirconstudios.com/Synth%20Example%201.flp

Example #2: Synth Bass

http://www.zirconstudios.com/Synth%20Example%202.mp3

Basic synth bass. Please note:

  • Only one oscillator in use ("Mix" knob all the way left). Simple can sometimes be better.
  • Square wave is being used because square waves are rich in lower harmonics. eg. nice for basses.
  • Similar envelope as the lead - we want a lot of sound coming out of this patch.
  • Lowpass filter with low cutoff with Velocity control. High frequencies = bad here, but we don't want it completely muffled.
  • Filter envelope I sort of just played with until I found the right settings. While I was editing it, I had the pattern playing.
  • EQ is set to boost up the low end freqs a bit.
  • Legato + Portamento for that fun gliding sound.

Example #3: Synth Pad

http://www.zirconstudios.com/Synth%20Example%203.mp3

This is a sort of subdued synth pad that you could easily use for harmony in a multitude of different song types.

  • Amplitude envelope has an extended attack, decreased decay/sustain, and extended release. This gives it a less abrupt sound, and more of a flowing one that sort of 'eases in'.
  • Filter is a LP12 so that it's not completely cutoff like a bass, but still somewhat subdued for that quieter, nice sound.
  • Delay settings; longer time, more obvious (higher Level). Gives it a better pad-like flavor.
  • 2 detuned saws for oscillators. This produces just a pleasant base sound, especially when you cut a chunk of it out.
  • Chorus/flanger - low level, but enough to produce some subtle variation in the sound.

Example #4: Short Synth

http://www.zirconstudios.com/Synth%20Example%204.mp3

You might use this if you have sonic space that needs to be filled, but don't want something as overbearing as a pad. A litte arpeggio/pluck type synth can help here.

  • Amplitude envelope is important. No sustain, no release, no attack, short decay = short sound.
  • Modulation envelope + FM is a result of just tweaking. I messed around with these settings until I found a sound I liked. I didn't really have anything in mind.
  • Filter is highpass because I didn't think this short synth should take up the sonic space of the bass. It could easily play some riffs like the one I have sequenced now and it wouldn't be at all intrusive.

End of Part 3

So, there you have it. Synthesis 101. I strongly encourage messing around with Synth1 and seeing what you can come up with - it's easy to use, and hopefully based on my example patches, you should have at least a vague idea of what type of settings can produce what type of sound.

Reference

DOWNLOAD SYNTH1 ! (free) --- http://www.zirconstudios.com/Synth1%20VST.dll Stick it in your FLStudio5/Plugins/VST folder, or whatever other VST folder you use.

-zircon

Promotion

Latest Albums

Latest ReMixes

Navigation