Requantise to 8-bit
Drop an audio file and export a WAV whose samples sit on the 8-bit integer grid and whose header declares that width. Triangular dither is applied before rounding.
256 steps, and you can hear them
An 8-bit sample has 256 possible values against the 65,536 of 16-bit. The theoretical noise floor lands near -50 dB, about 48 dB higher than 16-bit and comfortably inside what anyone can hear.
That is not a defect to work around at this depth. Anyone converting to 8-bit is doing it for the sound, and the audible floor is the sound.
Dither changes the character
Without dither, the quantisation error follows the signal and is heard as distortion that moves with the music, harsh and unstable in fades and tails. With triangular dither, the error becomes steady hiss and the signal underneath stays cleaner.
Both are legitimate at 8 bits. This page applies dither, which gives the more controlled result. The full tool turns it off if the rawer, more distorted version suits the material better.
Not quite a bitcrusher
A bitcrusher reduces effective resolution inside a normal file, usually alongside sample rate reduction, and the output is still 16 or 24-bit. This writes real 8-bit PCM: the file is genuinely 8-bit, the header says so, and anything reading it sees the reduced depth.
For the effect version, with resolution and rate controls that can move while the audio plays, use the Bitcrusher.
Where it fits
- Chiptune and retro production where the era is the aesthetic.
- Game audio matching a specific hardware generation.
- Embedded playback with hard memory limits.
Other depths
16-bit is the delivery standard and 24-bit the working depth. To change the dither type or add noise shaping, use the full Bit Depth Converter.