Understanding Sample Rates and Bit Depth Explained

Two Numbers Behind Every Recording

Open the settings panel of almost any audio interface and you will see the same two figures waiting for you: something like 48 kHz and 24-bit. They look modest enough, but between them they decide how faithfully a performance survives the journey from vibrating air to a waveform on a hard drive. Get them wrong and you either lose detail you can never recover, or you fill your storage with files far larger than the job required.

It helps to think of digital audio as a very fast series of measurements. The sample rate is how often we take a measurement; the bit depth is how precisely we write that measurement down. Everything else follows from those two ideas.

Sample Rate: How Often We Take a Snapshot

A sample rate of 44,100 Hz means the converter measures the incoming voltage 44,100 times every second. Each measurement becomes a number, and a stream of those numbers rebuilds the waveform on playback.

The rule that governs all of this is the Nyquist theorem: to capture a frequency accurately, you must sample at more than twice that frequency. Human hearing tops out around 20 kHz, so a little over 40,000 samples per second is theoretically enough. That is why 44.1 kHz became the compact disc standard, and why 48 kHz became the default for television, film and video work — it divides neatly into video frame rates.

• 44.1 kHz — music delivery, CD masters, streaming releases.
• 48 kHz — video, broadcast, podcasts and most studio recording.
• 96 kHz and above — heavy pitch shifting, time stretching, sound design and ultrasonic research.

Going higher does not make a recording sound "more analogue". What it does is give you extra bandwidth above the audible range and finer resolution when you slow audio down or transpose it. A bat call recorded at 192 kHz can be pitched down into hearing range; the same call captured at 44.1 kHz simply is not there.

Bit Depth: How Precisely We Measure Each Snapshot

If sample rate is the horizontal axis of a graph, bit depth is the vertical one. Each sample is stored as a binary number, and more bits means more possible values between silence and the loudest peak.

The maths is tidy and worth remembering: every extra bit adds roughly 6 dB of dynamic range. So 16-bit audio gives you about 96 dB between the quietest detail and the loudest peak, while 24-bit stretches that to around 144 dB — wider than any room, microphone or pair of ears can genuinely use.

The benefit of 24-bit recording is not louder peaks. It is headroom. You can track a drummer at peaks of −12 dBFS, safe from clipping, and still have a noise floor so low that it disappears entirely under the room's own ambience. That spare room is what makes 24-bit the sensible working format for anything you intend to edit.

What That Means for File Size

Uncompressed audio size is easy to estimate: sample rate × bit depth × number of channels. Running the numbers for a minute of stereo audio is revealing.

• 44.1 kHz / 16-bit stereo — roughly 10.6 MB per minute.
• 48 kHz / 24-bit stereo — roughly 17.3 MB per minute.
• 96 kHz / 24-bit stereo — roughly 34.6 MB per minute.

A three-hour multitrack session at 96 kHz quickly runs into tens of gigabytes before you have added a single plug-in. This is the trade-off in plain terms: higher values preserve more detail and generate bigger files to store, back up and send across the internet.

Choosing Settings for Real Projects

Matching the format to the job avoids both regret and wasted disk space.

• Podcasts and voice work: 48 kHz / 24-bit. Headroom for editing, straightforward delivery to video platforms.
• Music recording: 48 kHz / 24-bit in the session, then export at the delivery format the release requires.
• Sound design and effects libraries: 96 kHz / 24-bit, so extreme processing still has material to work with.
• Archival capture: the highest rate your converters handle honestly, with a backup of the originals untouched.

Where bit depth drops — say from 24-bit down to 16-bit for a CD master — apply dither. It sounds counter-intuitive, adding a whisper of noise to improve quality, but it decorrelates the quantisation error and keeps quiet fades and reverb tails sounding smooth rather than gritty.

Myths Worth Putting to Bed

Upsampling a 44.1 kHz file to 96 kHz does not restore information that was never captured. It changes the container, not the contents. Likewise, a 16-bit file converted to 24-bit gains no extra dynamic range; the noise floor travels with it.

The genuine gains come at the moment of capture, when the sound is still in the room. Record generously at 24-bit, pick a sample rate that suits your editing plans, and you will have a file that survives whatever you throw at it later — without filling an entire drive to get there.