Bit Depth Calculator

Bit depth fixes the amplitude resolution of a digital audio sample: the number of quantization levels available to describe the instantaneous voltage. N bits give 2 to the power N levels, so 16-bit audio has 65,536 steps and 24-bit has 16,777,216. Each added bit halves the size of the quantization step, and because a halving is a factor of two in amplitude it corresponds to 20 times the base-ten logarithm of 2, or 6.0206 dB. That constant is why dynamic range is bit depth multiplied by 6.02 - about 96 dB at 16 bits and 144 dB at 24 bits - and why the theoretical signal-to-noise ratio for a full-scale sine wave is 6.02N plus 1.76 dB, giving 98.09 dB at 16 bits. The 1.76 dB term comes from the RMS ratio between a sine wave and uniformly distributed quantization error. The practical benefit of 24-bit recording is headroom rather than audibility. Tracking at a conservative peak of -18 to -12 dBFS still leaves an enormous margin above the noise floor, so an unexpected transient does not clip, and 24-bit costs exactly 50 percent more storage than 16-bit. Two limits are worth stating plainly. The theoretical numbers assume ideal quantization, whereas the best professional converters achieve roughly 120 to 124 dB of real dynamic range, so the lowest bits of a 24-bit recording sit below the analog noise of the microphone and preamp. And reducing depth for delivery should be done once, with dither applied at that final step, to convert correlated quantization distortion into benign broadband noise.

Pulse-code modulation, the scheme that makes bit depth meaningful, was invented by Alec Reeves at the ITT laboratories in Paris in 1937 and patented in France in 1938 and Britain in 1939. It was first deployed for secure wartime speech in the SIGSALY system built by Bell Telephone Laboratories and used from 1943. Bernard Widrow at MIT published the foundational statistical analysis of quantization in the 1950s, establishing the model of quantization error as an additive noise source that underlies the 6.02N plus 1.76 dB expression. The 16-bit figure entered consumer audio through the Compact Disc. Sony and Philips agreed the Red Book specification in 1980, pairing 16-bit linear PCM with a 44.1 kHz sample rate, and the resulting 96 dB dynamic range comfortably exceeded that of the LP and cassette it replaced. Dither theory matured alongside it: work by John Vanderkooy and Stanley Lipshitz at the University of Waterloo in the 1980s showed that correctly applied triangular probability density dither decorrelates quantization error from the signal entirely, and noise-shaped dither systems from Sony, Apogee, and others followed in the early 1990s to move that noise into less audible bands. Professional recording moved to 24-bit as converter technology improved through the 1990s, and 32-bit floating point, using the IEEE 754 format standardized in 1985, became the normal internal representation for digital audio workstation mix buses because it cannot clip during intermediate summing.