Why this chapter matters
You are about to record a vocal. Your DAW's recording window asks for a sample rate and a bit depth, and your colleague shouts from the couch that you really should work at 192 kHz, because that sounds sharper. Later that week a videographer wants your track as an MP3, your distributor asks for a WAV, and somewhere along the way your DAW asks whether you want to apply dither. If you do not know what these terms mean, you guess. Sometimes that works out. Sometimes you deliver a clipped recording nobody can save anymore, or you archive years of work in a format that has quietly thrown information away.
This chapter gives you the mental model behind digital audio: how sound becomes numbers, what those numbers do and do not capture, and how to translate that into practical choices when recording, processing and exporting.
What you will learn
- describe A/D and D/A conversion and oversee the digital pipeline;
- explain sample rate and the Nyquist frequency correctly;
- understand bit depth and quantization noise and apply them when recording and exporting;
- frame 32-bit float correctly: what it does and does not solve;
- distinguish lossless and lossy formats and choose an archiving strategy;
- apply dither purposefully at the right moment in your workflow.
Who this chapter is for
For every producer who works with a DAW, from beginner to advanced. You need no mathematical background. Knowledge from the module on hearing and psychoacoustics is useful but not required: any terms you are missing can be looked up in the free intro lessons of that module.
A preview of the core idea
Digital audio is not a staircase of blocks. A common explanation shows samples as stair steps that supposedly make the sound angular. In reality the D/A converter reconstructs a smooth wave from those samples. Technical fact: for a signal that stays neatly below half the sample rate, the so-called Nyquist frequency, that reconstruction can be very accurate. At 44.1 kHz that limit sits at 22.05 kHz, above the limit of human hearing. Why you sometimes still choose a higher sample rate anyway, and when that mainly costs storage space, you will learn in the next lesson.