Bit Depth

In-Depth Explanation

Bit depth is the number of bits used to represent the amplitude of each audio sample during analog-to-digital conversion. It directly determines the dynamic range of a digital recording: the difference between the quietest and loudest sounds the system can capture before noise or clipping occurs.

How Bit Depth Works

When an audio interface converts analog sound into digital data, it takes rapid snapshots of the sound wave. The Sample Rate determines how often those snapshots are taken (the time axis). The bit depth determines how precisely each snapshot captures the volume (the amplitude axis).

A higher bit depth gives the converter more possible amplitude values to choose from. More values mean less rounding error. This rounding error produces a low-level noise called quantization noise, which sets the noise floor of the recording.

The math is straightforward. Every 1 bit of depth adds approximately 6 decibels (dB) of dynamic range:

• 16-bit audio: 96 dB of dynamic range, 65,536 possible amplitude values. This is the standard for CD-quality audio and most streaming services.
• 24-bit audio: 144 dB of dynamic range, 16,777,216 possible amplitude values. This is the professional standard for recording and mixing.
• 32-bit float audio: Over 1,500 dB of theoretical dynamic range. In a 32-bit float environment, internal clipping is mathematically impossible within the DAW's signal path.

Real-World Example

A singer records a vocal at 16-bit with the microphone gain set conservatively. The average recording level sits at -24 dBFS. The noise floor of the 16-bit recording is at -96 dBFS. The signal-to-noise ratio is 72 dB. If the singer suddenly sings 20 dB louder than expected, the signal peaks at -4 dBFS. No clipping occurs, but the quietest parts of the performance are only 72 dB above the noise floor.

The same singer records at 24-bit with identical gain settings. The noise floor drops to -144 dBFS. The signal-to-noise ratio is now 120 dB. The quietest breath is captured with far less noise. If the singer sings 30 dB louder than expected, the signal peaks at +6 dBFS. At 24-bit, this would clip. But if the recording was done at 32-bit float (now available on field recorders like the Zoom F2 and Sound Devices MixPre series), the overshoot is not destructive. You can simply lower the clip gain in your DAW and the unclipped audio reappears.

Why It Matters for Independent Artists

Always record at 24-bit. Set your audio interface and DAW to 24-bit before tracking. This gives you 144 dB of dynamic range, which means you can record at conservative levels (averaging -18 dBFS) without worrying about noise floor or clipping. Leave at least 6 dB of headroom on every input.

Mix and process in 32-bit float. Most modern DAWs (Ableton Live, Logic Pro, Pro Tools, FL Studio) handle internal processing at 32-bit float automatically. This gives you effectively infinite headroom on plugin chains and bus routing.

Export your final mastered release at 24-bit WAV. Most distributors (including DistroKid and CD Baby) accept 24-bit files directly. If you need to deliver a 16-bit file (for example, to burn a CD), apply dither during the final export. Dither adds low-level noise that masks the quantization error introduced by reducing the bit depth. Without dither, reducing 24-bit to 16-bit produces audible distortion on quiet passages and fade-outs.

For streaming platforms, 16-bit/44.1kHz remains the baseline delivery format. However, services like Apple Music, Tidal, and Amazon Music HD accept 24-bit files for their lossless and hi-res tiers. Read our guide on mastering for streaming platforms to understand the loudness and format requirements for each service.

Related Terms

• Sample Rate
• DAW (Digital Audio Workstation)
• Lossless
• LUFS
• Mastering