Bit Depth

In-Depth Explanation

What is Bit Depth in Audio?

To understand Bit Depth, you first have to understand how analog sound (continuous sound waves in the air) is converted into digital audio (ones and zeros inside a computer).

During the analog-to-digital conversion process, the audio interface takes "snapshots" of the sound wave.

• The Sample Rate determines how often those snapshots are taken (the horizontal axis of time).
• The Bit Depth determines how precise each of those snapshots is (the vertical axis of amplitude/volume).

Bit depth is simply the number of bits of information used to represent the amplitude of the audio signal at any given moment. The higher the bit depth, the more possible amplitude values can be recorded, which translates directly to the dynamic range of the audio file.

Dynamic Range and the Noise Floor

Dynamic range is the difference in volume between the loudest possible peak in a recording (before it distorts) and the quietest possible sound (the noise floor).

When audio is digitized, the continuous analog wave is "rounded" to the nearest available digital value. This rounding process creates a very quiet form of distortion called "quantization error," which we hear as a low-level hiss or white noise. This is the noise floor.

A higher bit depth gives the computer more "rungs on the ladder" to choose from when rounding the analog signal. More rungs mean less rounding error, which means a lower noise floor and a larger dynamic range.

The Math of Bit Depth

For every 1 bit of depth, you gain approximately 6 decibels (dB) of dynamic range.

• 16-bit audio: (16 x 6 dB) = 96 dB of dynamic range. It has 65,536 possible amplitude values.
• 24-bit audio: (24 x 6 dB) = 144 dB of dynamic range. It has 16,777,216 possible amplitude values.
• 32-bit float audio: Provides a theoretical dynamic range of over 1,500 dB, which is greater than the difference between absolute silence and a nuclear explosion.

16-bit vs. 24-bit: Which Should You Use?

16-Bit: The Distribution Standard

16-bit audio (with a 44.1 kHz sample rate) has been the standard for consumer audio since the invention of the Compact Disc (CD) in the early 1980s. A dynamic range of 96 dB is more than enough for human listening environments. For context, the background noise in a quiet living room is around 30-40 dB SPL, and sounds above 120 dB SPL cause physical pain. Therefore, a 96 dB range comfortably covers everything from a whisper to a scream in a typical listening environment.

Almost all streaming services (Spotify, Apple Music) stream audio at 16-bit/44.1kHz (though some offer 24-bit "Hi-Res" tiers).

24-Bit: The Recording Standard

While 16-bit is fine for listening, 24-bit is the absolute standard for recording and mixing.

Why? Because when you record in 16-bit, you have to record very close to the maximum volume (0 dBFS) to keep the quiet parts of the signal far above the noise floor. If a singer suddenly screams, the signal will clip and distort permanently.

Recording at 24-bit pushes the noise floor down so far (-144 dB) that you can record at a very conservative, quiet level (averaging around -18 dBFS) and completely eliminate the risk of digital clipping. You can then boost the volume later during mixing and mastering without ever bringing the noise floor up to audible levels.

What about 32-Bit Float?

Many modern DAWs and field recorders use 32-bit floating-point processing. In a 32-bit float environment, it is mathematically impossible to permanently clip the audio internally. If a signal goes "into the red" on a 32-bit float channel, the data isn't destroyed; you can simply turn the fader down, and the un-distorted audio will reappear.

Best Practices for Producers

1. Always Record at 24-bit: Set your DAW and audio interface to 24-bit. This allows you to leave plenty of headroom when tracking.
2. Mix in 32-bit Float: Most modern DAWs do this automatically, giving you infinite headroom on your plugins and master bus.
3. Export Final Mixes at 24-bit: Deliver your final, unmastered mixes to the mastering engineer as 24-bit WAV files.
4. Dither Down to 16-bit for Final Release: If you must bounce a 24-bit file down to a 16-bit file (e.g., to burn a CD), you must use a process called "dithering" to mask the quantization noise caused by throwing away 8 bits of data. However, many modern distributors (like DistroKid) now accept 24-bit files directly and will handle any necessary down-conversion themselves.