What is dither? When you use it?

Topics like loudness, compression and mastering are the main topics that interest all engineers.

Dithering can be the subject of a slightly inaccessible topic, as it has a lot to do with the basics of digital audio.

However, applying the screening correctly is very simple and easy. A solid understanding of the “what to do” and “why to do” parts will eliminate the need to worry about hesitation in the future.

What is dithering?
Dithering is the addition of low-level noise to the audio to compensate for loss and reduce errors (noise) when changing the bit depth.

You may be feeling, “Why does adding noise make my recordings sound better? Noise is bad, isn’t it?”

Dither is not just a musical term, and the noise you add to complement it with dithering is actually similar to random fluctuations.

In fact, the word dither means “neural oscillation.” Its meaning comes from the way dither was discovered.

Engineers have found that computers in mechanical aircraft work with greater precision in flight than on the ground.

The vibrations from the aircraft engine actually improved the precision of the viscous moving parts of the machine.

The noise that dithering adds to the track works the same way. Helps improve the accuracy of digital audio files.

When should I use dithering?
Before we dive into dithering, let’s start with the best practices.

If you follow these three guidelines, you shouldn’t have a problem using dithering.

Never change the type of file you want to make unless necessary. If you’re tracking at 24-bit / 44.1 kHz, stick with that setting! If for some reason you need to lower the resolution, be sure to blur during the conversion.
Save the screening until the file is exported from your DAW. Dither only once during export.
If you are submitting the file for mastering, don’t hesitate if you can export the 32-bit floating file. In this case, mastering takes care of the dithering. When exporting anything other than 32-bit float, you must interpolate. This includes bounce files with the same bit depth as the recorded bit depth.
That is all! If you don’t want to think about dithering again, stop here and go back to your last project. But if you don’t feel familiar yet, or want to know why you should follow these rules, read on.

Here’s everything you wanted to know about dithering, but was afraid to ask.

When you select the bit depth, the precision of the file that you actually create during recording is selected.
Bit depth
Let’s start with the basics. Let’s talk about digital audio here!

A file created by a computer during recording is a digital representation of an analog signal connected to an audio interface.

Before you start recording, you need to select the sample rate and bit depth for your DAW session. Choosing a bit depth selects the precision of the file you actually create during recording.

Each bit added improves the resolution at which the ADC (analog / digital converter) needs to measure the level of the input signal.

Therefore, we recommend burning a 24-bit file for LANDR mastering. 24-bit is the best option available in most DAWs.

Reduce the impact of mistakes
Once the file is in your DAW, be careful when lowering the resolution. If you are not careful, reducing the bit depth will result in a digital error.

However, at some point in the process you need to lower the resolution. The reason is as follows.

DAWs perform calculations that boost faders, buses, and plug-ins at higher resolutions than audio files.

DAWs perform calculations that boost faders, buses, and plug-ins at higher resolutions than audio files.
DAWs perform calculations that boost faders, buses, and plug-ins at higher resolutions than audio files.

This means that unless you export the file in 32-bit float, the resolution of the track will be lowered and you will get an error when exporting from your DAW.

Digital file errors cause problems that are directly related to the way real-world tracks sound, such as noise and distortion.

This distortion occurs only in the quietest part of the system’s dynamic range, but is corrected with proper blurring.