Basics of Synthesizers (7) – Granular Synthesis

g

Hi! This is Jooyoung Kim, a mixing engineer and music producer. Today, I’ll introduce granular synthesis, which is composed of “grain” units.

The principle of this type of sound synthesis is very similar to the concept of sampling. Granular synthesizers take small parts of a sample and store them as units. Those units are called “grains.” That is where the name “Granular Synthesis” comes from.

The grains have lengths ranging from approximately 1 to 100 ms, and these samples can be played at different times, phases, speeds, and frequencies in a granular synthesizer. Thus, cloudy sounds that resemble long reverberations and continuous tails are produced by these manipulations.

In 1947, Dennis Gabor introduced this concept, which was first implemented in a computer in 1974 by Curtis Roads, an electronic music composer specializing in granular and pulsar synthesis. The spans are quite long, aren’t they?

The emergence of granular synthesizer hardware was significantly delayed due to the large computational requirements and high CPU speed needed for the calculations. Thus, the real-time version of this synthesizer emerged 12 years after the use of granular synthesizers in Curtis Roads’ music.

The real-time granular synthesis was invented by Barry Truax, a Canadian composer. This synthesis was processed by the DMX-1000 computer.

Naturally, you might wonder about the existence of physical hardware, as with other synthesizers. The upper photo shows the modern, dedicated granular synthesizer hardware, the GR-1, which was released in 2017.

This shows that this kind of synthesis processing requires a lot of computing resources. And this means that many virtual synthesizers using this method require significant computing power.

However, for implementation in DAWs (Digital Audio Workstations), no physical hardware is needed to use granular synthesis. Arturia’s EFX Fragments is a great example.

Output’s “Portal” plugin is another excellent granular effect.

This is the official video of the EFX Fragments. You might hear the “shimmering sound” from it.

This sound could be effective in ambient music that requires long tails on the sounds. I like these sounds, but I’ve never used them in my music… ^^;; However, someday I will…

OK, I’ll wrap this up today. See you in the next post!

Basics of Mixing – 6.5 Decompressor, Expander, Gate

Hello everyone, this is Jooyoung Kim, an audio engineer and music producer.

Today, I’d like to talk about some friends that, like the compressor, help us control dynamics: the Decompressor, Expander, and Gate.

Decompressor

A decompressor works in the opposite way to a compressor, making sounds above a set threshold louder.

You might wonder, “Why would anyone use this?” Occasionally, it can be used to force a groove or to emphasize transient parts.

Expander and Gate

An expander decreases the volume of sounds below a certain threshold. When the ratio is increased, it starts to function as a gate.

Think of the relationship between an expander and a gate as similar to that between a compressor and a limiter. The parameters for these tools are similar to those of a compressor. If you understand how a compressor works, you can easily use these tools as well.

These effects can be particularly fun when used with sidechain.

For example, you can apply a gate to a track with white noise and use sidechain so that the white noise only plays when the hi-hats come in. Or, you can apply a decompressor to a track and use sidechain to make the instruments louder only when the snare hits.

There are countless creative ways to use these effects if you let your imagination run wild.

If you’re a songwriter, try applying these effects not only to your mixes but also to your instruments. They can be very effective.

That’s all for today. In the next post, we will look at other effects that have compressor-like functions. Stay tuned!