The options available to artists with modern recording software are almost infinite. A huge aspect of this software is its effects. They manipulate how an audio signal sounds to produce the desired outcome. Although lesser-known, dithering is one of these effects. In this article, we will cover what is dithering and how it is used in music production.
Dithering is something you may well never of heard of, never mind used. Because its use and impacts are quite complicated, musicians tend to shy away from using it. Despite this, once familiar, it can be an extremely useful tool that can benefit a track greatly. It shouldn’t be avoided, but rather embraced as a significant resource.
First, we will cover what is dithering and describe its impact on music. Then, we will tell you how to use it for your own music and the different types of available dither.
Dither is added to songs to remove unwanted distortion. The way in which it does this is quite peculiar, by adding noise to the track. This might seem strange at first, but you trade the unwanted distortion for the noise added by dithering. A very low-level noise is included, which cancels out the distortion when listening to the song.
When producing a track, you most likely work with the highest possible resolution, 24-bits for example. However, playback options often don’t support this higher resolution, meaning it needs to be reduced when mastering the track. This is especially true for CDs, only supporting 16-bits resolution. We all know this reduction in resolution can distort the track adding undesired sounds. Dithering audio adds a low-level hiss which, to the human ear, cancels out the distortion.
This process is known as dithering signal processing.
We’re now going to explain the actual technical side of how dither works.
This is quite complex and you certainly do not need to understand it to use dither. Still, a lot of you will likely want to understand the process you are adding to your tracks.
First, we need to explain bit-depth.
This defines the number of measurement values to describe the amplitude of a single audio sample. Each bit then represents 6db of dynamic range. So, a 24 bit-resolution track will have a potential range of 144db (24×6). The image above should give you a visual indication of what we mean here.
This fundamentally offers lots of available values to describe the dynamic range of your track. In other words, it is more specific.
Distortion can occur when trying to reduce the resolution of a track. As we mentioned earlier, this is necessary when sharing songs on platforms and storing them on devices that only support a lower playback.
Reducing the resolution, reduced the number of values available to measure the dynamic range. For example, you’re reducing your track from 24-bits to 16-bits resolution. You now lose 8 available values to measure the dynamic range.
This results in values being rounded off to meet the new numbers of bits and can cause distortion and the loss of very low-level signals. This is known as truncation distortion.
When adding dither before reducing the bit-depth, you basically trade this truncation distortion for this low-level noise. You counter the lost quality by adding something else to be altered rather than the track itself.
After that, as a music producer, you’re probably interested in using this effect. You may even be wondering why you haven’t used this technique before with quantization distortion occurring so often. Now we will discuss when to use dithering.
There is only one time you should ever be applying dither to your track; during mastering. Never should you add noise or undesired sounds to a track during the production process. You should want to produce the best quality track possible, adding noise should only be a necessary step after production has finished.
You also never want to add dither unless it is absolutely necessary. This should be obvious, but you generally never want to add anything to your track apart from the desired sounds. Dithering should be a last resort when changing the resolution has distorted your track to an uncomfortable level. Distortion is almost impossible to avoid, it’s up to you to judge whether it’s severe enough to add dither.
Finally, never use dithering on a single track more than once. It should only ever be added before you alter the bit-depth after the track is finished.
Different types of dither offer noise shaping. This allows you to add an eq curve to the dither noise, which moves the energy of the noise to less audible regions within the frequency spectrum. Basically, it removed distortion still but adds the noise to where it’s least likely to be heard. The dithering audio mastering you use depends on the type of track you are producing.
This is the most common form of dithering and is usually the default. This producer a lower peak amplitude of noise than shape, but is concentrated less in the high frequencies. This is the safest method if any additional processing of the track might occur. If applied twice, it won’t necessarily be as degrading if shaping was added twice.
Rectangular is very similar to triangular in the noise they make. Instead of adding it to the lower frequencies, randomization is used to distribute the noise evenly. This does not add noise to absolute silence and approximates white noise.
There are different POW-r modes that offer a higher level of dither, but add it above the audible range. This is true for all the different types of dither, but especially POW-r modes, only apply dither if the track we undergo no further processing. This is essential to not destroy the quality of your track.
This is best used for example, on compressed pop music. It is optimized for lower dynamic range programs.
This is best for less complex programs like spoken word.
This is targeted at complex, highly dynamic range programs. For example, orchestral performances.
We hope after reading this you are feeling more familiar with dithering. The concept behind the process is quite complicated, but its impacts on a track aren’t too hard to understand. If you think of it as trading distortion for inaudible noise, you should have a basic understanding. It is better to have inaudible low-level noise than audible distortion.
Distortion when reducing bit-depth is something many musicians struggle with. You are always going to produce a track at the highest bit-depth, being the highest quality. Unfortunately, this size of bit-depth will never be accepted for playbacks. Almost always you will have to reduce the resolution to 16-bits. Using dither to counter truncation distortion is a powerful tool.
We know that was a lot of complex information. Here is a quick summary that includes the main points: Dither reduces truncation distortion by trade this distortion with low-level noise. It should only ever be applied once, during mastering the audio file before you decrease the resolution. If the resolution is staying constant, there is no need to apply dither.
Producing tracks to the highest possible quality is fundamental for a musician. Don’t let the thought of distortion cause you to compromise on quality. Dither is here exactly for that reason.
Get free music distribution and find opportunities to get your music in film, TV, and more through sync licensing. Finally, you can amplify your music to those that need to hear it music promotion and professional sharing tool. Try all of this out for yourself by joining Music Gateway. Get your free trial, no strings attached.