We live in a digital age. While we still use analog devices here and there, digital technology dominates our lives. Today, the recording industry has mostly abandoned analog methods and transitioned strictly to digital audio for the majority of production purposes. We’ve gone over the differences between analog and digital recording methods, but let’s dive deeper into the machinations of digital audio, specifically regarding bit depth.
Digital Bits: Cracking the Binary Code
To understand bit depth, you must first have a grasp on digital audio. Recall that an analog sound wave is a continuous movement of energy passed through a medium (typically air). The key word here is continuous. Digital audio takes that continuous information and converts it into fragmented or binary data (1s and 0s) called samples. Remember that the word digital contains the root “digit,” or numerical unit. In other words, when recording digitally, the once fluid sound wave becomes a series of digital audio samples, coded by groups of 1s and 0s to replicate said sound wave.
Imagine an archway (analog sound), then imagine a staircase transposed over that arch (digital). The stairs must follow the same path and reach the same height (amplitude) as the arch, but you must jump from one step to the next to climb it (quantization). The more stairs, the closer you are to replicating the analog sound wave; the fewer stairs, and you take bigger, blockier leaps to scale it. You have to round up or down depending on which step you’re closer to (rounding error). These steps represent the rigid binary code, or the 1s and 0s. You might wonder, then, how many 1s and 0s make up one of these samples. The answer lies with bit depth!
What is Bit Depth?
We can think of bit depth as the resolution for each given digital sample. Each binary digit is called a bit. The bit depth will determine how precisely we can define the amplitude. For example: If we only had 1 bit, the only 2 possible definitions would be that the amplitude is 1 or 0, on or off. With more bits, we can add more precise values for the amplitude that accurately sample the original analog signal. The most common bit depths you’ll come across are 16-bit, 24-bit, and 32-bit float. You might guess that the higher the number, the more headroom and dynamic range you have, and you’d be correct. However, the difference between 16, 24, and 32-bit float is more vast than you might think.
A 16-bit depth means that an audio sample’s amplitude can be a 16-digit number made of 1s and 0s (could be 0000000000000000, 1111111111111111, etc.). If you do the math, this means that a 16-bit depth gives you 65,536 potential values for just one audio sample. Do the same math for a 24-bit depth and you now have 16,777,216 potential values. That’s 256 times more! If you’re wondering where 32-bit float comes in, “float” refers to a variation in spacing between integers. With typical bit depths, each value is evenly spaced (remember the staircase). The steps of a 32-bit float depth, however, change in proportion to the audio signal. This results in bigger rounding errors for larger float values but smaller errors for lower values. It also means that you won’t encounter clipping when in 32-bit float due to the floating point.
How Does Bit Depth Affect Recording and Mastering?
Okay, that’s enough math and theory for now. We care more about the practical applications of using different bit depths, after all. So which bit depth is best for recording? We have to keep a few things in mind: audio quality (or resolution), data size, and playback format. The higher your bit depth, the higher resolution your digital audio will have. So, why not record everything in 32- or even 64-bit float?
For one thing, higher resolution means larger file sizes and slightly increased CPU usage, but not enough to make a noticeable difference when recording. In truth, recording at 32- or 64-bit float isn’t all that necessary anyway unless you want some additional headroom. When sound waves are converted from analog to digital, they enter the DAW at a 24-bit depth by default. Bit depths above this value only matter once you introduce additional processing. For many producers, 24 bits is plenty of dynamic range, but some will choose 32-bit float or more just in case.
Lastly, consider that mastering and exporting your mix will require bit reduction. To save space, playback formats (like CD) often stick to 16 bits. So, anything you record above this resolution will be crammed to fit a smaller space. Techniques such as dithering can help maintain the dynamics of a mix during this conversion process. This doesn’t mean you should just record everything at 16 bits, however. Doing so will give you fewer options when recording and mixing. Even if you lose some of the quality after exporting your mastered track, working in a higher resolution will enhance the overall quality of your digital tracks.
To sum up, bit depth refers to the amount of digital information that can be stored in an audio sample. The higher the bit depth, the higher the resolution, meaning the digital audio can better replicate the raw analog sound wave. In most cases, recording in 24 bits will get you where you want to go, though higher bit depths will yield increased headroom and more potential dynamic range. In the end, your master will most likely come out at 16 bits so it’s optimized for all playback formats. MasteringBOX automatically renders your tracks in 16 bits when exporting in WAV (24bit in HD). Now that you know a bit more about bit depth, you can record, mix, and export with confidence!