Why do the two controls sound different?

Because they degrade different things. Fewer bits quantises the level of each sample, which adds a steady hiss of quantisation noise, while a longer hold repeats each sample across time, which discards detail and folds high frequencies down as aliasing. One is a noise floor and the other is a moving shimmer.

What does the effective sample rate mean?

It is the source rate divided by the hold factor, and it describes the rate the result behaves like. A hold of four on a forty four thousand one hundred hertz file gives about eleven kilohertz, which is why the same hold factor sounds different on files at different rates.

Is the effect ever transparent?

Yes, at sixteen bits with a hold of one. Nothing is quantised further than the file already was and nothing is held, so the output matches the input. That is useful as a reference point when judging how much of a character a particular setting adds.

Why is no dither applied?

Because dither exists to make quantisation noise less objectionable when the aim is fidelity, and this tool's aim is the opposite. Leaving the noise correlated with the signal is what gives the effect its characteristic grain rather than a smooth, dithered refinement.

Can I crush only part of a file?

The whole file is processed in one pass. To crush a section, keep that section with the trimmer, crush it, and join it back into place with the joiner, which is the usual way to apply any of these effects to part of a recording.

Bitcrusher Effect

You Might Also Need

Distortion Effect

Add controlled audio distortion

Ring Modulator

Create metallic ring mod effects

Radio Effect

Simulate AM/FM radio sound

Sample Rate Calculator

Calculate sample rate requirements

Bitcrusher Effect reduces the resolution of a recording in two ways at once, which is how the sound of early digital equipment is imitated. Bit depth sets the coarseness of the levels and a sample-and-hold sets the coarseness of time.

What is Bitcrusher Effect?

Choose a file, set the bit depth and the hold, and crush it; the report gives the settings and the effective sample rate the result simulates.

The two controls:

A bit depth from two to sixteen, defaulting to eight, which is the number of levels each sample is snapped to.
A sample-and-hold from one to forty, defaulting to four, where the figure is how many input samples each output value is held for.
A report giving the bit depth, the hold factor, and the effective sample rate that the hold produces, which is the source rate divided by the hold.
The knowledge that sixteen bits with a hold of one is transparent, which means the effect has a genuine bypass setting rather than a near-miss.

The two reductions are independent and produce different artefacts, which is why both appear in one tool. Quantisation snaps each sample to a grid of levels; with fewer bits the step between adjacent levels grows, and the difference between the true value and the snapped one is heard as a hiss of noise that is present even during quiet passages. That is quantisation noise, and its level relative to the signal is set by the bit depth alone. The sample-and-hold works in time rather than in level: one sample is captured and repeated for several successive output positions, which discards the detail between those positions and folds any content above the resulting effective rate back down into the audible band as aliasing. The two artefacts therefore have completely different characters — one is a steady hiss and the other is a metallic shimmer that moves with the material — and combining them at moderate settings produces the familiar eight bit sampler sound of the mid nineteen eighties. The effective rate line exists because the hold is expressed as a multiple rather than as a rate, and the two are not equivalent: a hold of four on a forty eight thousand hertz file gives twelve kilohertz, while the same hold on a ninety six thousand hertz file gives twenty four, so the hold has to be read against the source rate to know what the result actually sounds like.

What it cannot do:

No dithering, so the quantisation noise is correlated with the signal rather than randomised.
No anti-aliasing filter, so the alias content from the hold is deliberately retained.
No separate control of where in each held block the sample is captured.
No reconstruction filter, so the output is a stepped waveform rather than a band-limited one.
No wet and dry mix.
Output is written as sixteen bit WAV while sounding far coarser.

Use it when the target is a lo-fi texture rather than a smooth saturation: eight bit drum machine grit, video game percussion, or the telephone era character of early digital recordings. When the aim is harmonics rather than resolution, the distortion effect produces a smooth saturation instead, and when the target is a specific recording format's limitations, the sample rate calculator and the bit depth calculator describe what those formats actually provide.

How to use Bitcrusher Effect

Drop an audio file onto the panel and let it decode.
Set the bit depth, starting at eight for a recognisable lo-fi character.
Set the sample-and-hold factor and note the effective sample rate it produces.
Crush the file and download it, or set sixteen bits and a hold of one to confirm the bypass.

When to use Bitcrusher Effect vs related tools

Bitcrusher Effect is the tool when the goal is deliberate loss of resolution rather than harmonic saturation. Distortion Effect adds harmonics with a smooth curve instead of quantising. Ring Modulator changes the spectrum by multiplication, and Radio Effect narrows the band to imitate a transmission rather than reducing the resolution.

Privacy & Security

This tool runs entirely in your browser — no data ever leaves your device. There is no server round-trip, no upload, no logging, and no account required. Your input is processed locally using client-side JavaScript and is never stored, transmitted, or accessible to anyone else. When you close the tab, everything disappears.