Audio Destruction: What is a Bitcrusher and How Does It Work?

What is a Bitcrusher?

A bitcrusher is an audio effect that deliberately degrades the quality of a sound. But why would you want to destroy your sound quality, you ask? Because if done tastefully, crushing your sounds is an excellent way to add dirt and grit, as well as emulate the fabled electronic instruments and samplers of old.

Audio production technology has become insanely good and unimaginably cheap in recent times, as compared to earlier centuries, and we enjoy the benefits of these trends. The bar to entry for music production has dropped so low, that even the average person who buys a MacBook has access to excellent music production software in GarageBand (the “lite” version of the very popular DAW, Logic Pro). It’s a well known fact that Finneas O’Connell produced Grammy winning album, When We Fall Asleep, Where Do We Go? for his sister, Billie Eilish, in a bedroom.

All that quality is well and good, but something gets lost in the process. If everything sounds too good – particularly in pop, rock, or electronic genres – it winds up sounding sterile and lifeless. Back in the day, the technology wasn’t really that great.

And that is what gave us the “sound” of the music from those eras. The Beatles wouldn’t sound like The Beatles if they were recorded in pristine digital quality. Likewise, Kraftwerk or Prodigy wouldn’t sound like themselves without the early electronic instruments they used to create the music that made them famous.

So what do we do when we want to capture that “sound” and apply it to our own contemporary works?

We use the amazing technology we have at our disposal to purposely color, degrade and destroy our audio. Bitcrushers are just one of the many tools at our disposal to do just that.

How Does a Bitcrusher Work?

Bitcrushers work by degrading the signal to which it is applied in two ways: bit and sample rate reduction. In order to better understand these processes, let’s review the concept of sampling as it pertains to the digitization of audio:

The sounds we hear in the world are “analog” sounds. They are real, and the sound waves that carry them exist in time – a continuum which can be infinitely subdivided. When these sounds are digitized in the recording process, the computer creates a representation of this continuum by periodically taking samples of the wave it is recording. This means the computer records the amplitude of the sound wave at regular intervals. These intervals occur tens of thousands of times per second, and when all of those tiny samples are heard in sequence, the recorded sound is heard.

Sample Rate Reduction

Also called “Downsampling,” this process reduces the number of samples taken over one second of time. Digital sounds of excellent quality use high sample rates to create more detailed sounds. To put this in context, a typical CD quality .wav file has a sample rate of 44.1 kHz. This translates to 44,100 samples per second. Some audio interfaces and DAWs can operate with sample rates of up to 192 kHz. That is some high resolution audio!

Downsampling a sound reduces the number of samples used to recreate the sound, deteriorating the quality and adding a digital distortion. Older samplers and DA/AD (D=Digital; A=Analog) converters simply did not function at higher sample rates. So, to replicate the sound and feel of a classic sampler, downsampling is in order. Often, this adds the color, dirt, and even girth to the sound that you may be looking for.

Bit Reduction

If we imagine a sound wave on a graph, sample rate manipulation is adjusting the number of samples taken over time – in other words, we are working with the X-axis. Bit reduction, often denoted as bit depth, is working with how many values are available to be assigned to a given sample – working with the Y-axis.

At a high bit depth, there are many values for a given sample to be. Let’s say, for example, we have the values of -10 to 10, with each whole number (-10, -9, -8, etc.) representing a possible value for a sample. That is 21 possible values (including 0). If we use a bitcrusher to reduce the bit depth, we lower the number of possible values available. So let’s say we cut the bit depth in half. That means we still have the values from -10 to 10, but we only have half the number of possible values in that range (-10, -8, -6, etc.).

So in both methods, we are reducing the resolution of the digitized sound. One over time, the other over depth. This creates distortion, as well as artifacts. An artifact is an aberration in the sound due to imperfections in the digital representation of the sound. The artifact we run into in regards to sampling is called aliasing.

Aliasing

Aliasing occurs when the DA/AD converter misinterprets frequencies in the sound it is sampling. There is a principle in audio called the Nyquist Theorem. The Nyquist Theorem states that the highest frequency that can be recorded by a digital system is equal to half of the sample rate. This means that if you are recording with a sample rate of 48 kHz, the highest possible frequency you can record is 24kHz, which is about 4 kHz higher than the human ear is capable of perceiving.

This is all well and good, but when you begin manipulating the sample rate of a sound by downsampling it, you begin to run into situations where the highest reproducible sound is within the range of hearing. When this happens, your DA/AD converter mistakes a high frequency for one of its lower frequency multiples. This is usually viewed as a bad thing, as it is a glaring flaw in the representation of the sound being recorded. But for our purposes – purposely degrading and manipulating the sound quality to achieve a desired effect – aliasing could something we are actively looking for. At the same time, knowing about aliasing and how it occurs allows us to avoid it if it is undesirable.

In Conclusion

Bitcrushing is an awesome tool for any producer to have in their pocket. It serves an aesthetic purpose that can be applied to any production, but it must be used judiciously and sparingly. Really fantastic, detailed sounds can sound even better when juxtaposed with sounds of a lesser resolution and quality, and lower quality sounds can often have a great impact. But misuse this tool, and your production will just sound low-quality and bad.

So when it comes to the bitcrusher, it’s best to know what you’re doing with it, how it works, and why it sounds the way it does.

Until next time,
AudioMunk

Want to learn a bit more about bit rates? Check out this previous article:
Bit Rates Explained for Digital DJs