Talk:Sample Format - Bit Depth

Bit Depth versus Sample Format

The more common term among computer musicians / audio technicians is "bit depth". The more common term for audio programmers is "sample format".

The terms have subtly different meanings.

• Bit depth: The number of bits (binary digits) per sample.
• Sample format: The "numeric type" of samples.

To illustrate the difference, "32-bit" may be "integer" or "floating point". Both could be described as having a "bit depth" of 32, but "32-bit float" is a different "format" (a different numeric type) to "32-bit integer".

The term "bit depth" is usually used in relation to "integer format linear PCM". In other words, The full amplitude range of sample values is divided into a large number of _equally_spaced_ values, where the number of distinct values is determined by the number of bits per sample.

The term "bit depth" is somewhat lacking when considering floating point formats, and/or non-linear amplitude scales (such as ADPCM), because the "depth" (distance from one value to the next) varies according to amplitude.