6 min read
Sample rate vs bit depth: which one actually changes how your film sounds
Two numbers, two completely different jobs. Confusing them is how projects end up recorded at 96 kHz and 16 bit.
Every recorder asks for both, and most people set them once and forget. They control different things, and only one of them is worth spending storage on for dialogue work.
Sample rate is horizontal
It counts snapshots per second and therefore sets the highest frequency you can capture, roughly half the rate. At 48 kHz you get up to about 24 kHz, past the top of human hearing. Higher rates matter mainly when you plan extreme pitch shifting for sound design, or when a spec demands them.
Bit depth is vertical
It sets how finely each snapshot is measured, which defines the dynamic range and where the noise floor sits. 16 bit gives about 96 dB, 24 bit about 144 dB. That headroom is what lets a quiet whisper survive a later level lift without hiss coming with it.
What to record
- Dialogue and general production sound: 48 kHz, 24 bit.
- Sound effects you intend to pitch down hard: 96 kHz, 24 bit.
- Anything destined for a broadcast spec: read the spec first, it usually says 48 kHz, 24 bit.
- Never 16 bit on set. The storage saving is trivial, the lost headroom is not.
What to deliver
Match the video project. 48 kHz is the video standard because it divides cleanly into frame rates. Delivering 44.1 kHz material into a 48 kHz timeline forces a resample somewhere, and the somewhere is usually a background process you cannot audition.
Converting without damage
When you do need a conversion, do it once, from the highest quality source, with a good resampler, and listen to the result. The converter here shows source and target values side by side, keeps the metadata visible, and runs locally in your browser.