Reading 32-bit colour as 32 bits per channel, when in most consumer software it means 8 bits each across four channels.
Colour
Bit depth
Also called colour depth, bits per channel or 8-bit vs 16-bit. Here is what it means, when it changes what you export, and what people get wrong.
ColourRuns on your deviceUpdated 10 August 2026
The short answer
Bit depth is how many bits an image uses to store each colour channel, which sets how many distinct levels that channel can hold. Eight bits gives 256 levels per channel, ten bits gives 1,024 and sixteen bits gives 65,536. More levels mean smoother gradients and more room to edit.
Updated
The short version
Bit depth at a glance
| Quick fact | Detail |
|---|---|
| 8 bits per channel | 256 levels, about 16.7 million colours |
| 16 bits per channel | 65,536 levels, for editing headroom |
| 24-bit colour | Per pixel: 8 bits each for R, G and B |
| 32-bit colour | Usually 8 bits each for R, G, B and alpha |
| Baseline JPEG | 8 bits per channel only |
| Main symptom of too few bits | Banding in smooth gradients |
Slide the table sideways to see every column.
What do the extra bits actually buy you?
| Bits per channel | Levels per channel | Colours in RGB | Where you meet it |
|---|---|---|---|
| 8 | 256 | About 16.7 million | JPEG, most PNG, everything on the web |
| 10 | 1,024 | About 1.07 billion | HDR content, AVIF and HEIC |
| 12 | 4,096 | About 68.7 billion | AVIF, cinema, many camera RAW files |
| 16 | 65,536 | More than the eye can distinguish | PNG, TIFF and PSD editing masters |
The point of the higher depths is not that you can see more colours in the final image. It is that you have levels to spend while editing. Every curve, exposure change and white balance correction stretches some part of the range, and stretching leaves gaps between the values that survive.
Why do 24-bit and 32-bit mean something different?
Because those numbers count per pixel, not per channel, and the labels are used inconsistently.
- 24-bit colour is 8 bits each for red, green and blue. It is the same thing as 8 bits per channel.
- 32-bit usually means 8 bits each for red, green, blue and alpha, so an 8-bit RGBA image. In visual effects and rendering, though, 32-bit sometimes means 32-bit floating point per channel, which is enormously more.
- 48-bit is 16 bits per channel across three channels.
When a specification is ambiguous, check whether the number divides neatly by three or four. If it does, it is almost certainly a per-pixel figure.
When does 8-bit start to show?
In smooth gradients, and it shows as banding: visible steps where a soft transition should be. A studio backdrop fading from lit to shadow, a sky, a soft vignette and the gradient behind a cut-out product are all classic cases.
The mechanism is straightforward. Suppose a gradient occupies only levels 90 to 120 in your original, thirty steps. Stretch it across the full range for contrast and those thirty steps still number thirty; they are just spread further apart, so the eye now sees each one as a band. A 16-bit file with thousands of levels in the same region has values to fill the gaps.
The habit that fixes it: edit in 16 bits, do all your tonal work there, and convert to 8 bits once at export. Converting late costs nothing. Editing an 8-bit JPEG through several rounds of correction, saving in between, cannot be undone.
Which formats support which depths?
PNG supports 8 and 16 bits per channel for full-colour images, plus lower depths for greyscale and palette images. TIFF handles 8, 16 and 32-bit floating point. AVIF supports 8, 10 and 12 bits. HEIC covers 8 and 10. Camera RAW files typically record 12 or 14 bits from the sensor.
Baseline JPEG is the outlier: 8 bits per channel, full stop. That is not a problem for a finished listing photo, since 8 bits displays perfectly well once the editing is done. It is a problem if the JPEG is your working master and you plan to keep correcting it.
What people get wrong about bit depth
Each one is a real failure mode, not a style preference.
Doing heavy curves and white balance work on an 8-bit JPEG, which spreads the surviving levels apart and bands the smooth areas.
Exporting listing photos at 16 bits per channel, which doubles the file size for detail no browser or buyer will ever see.
Assuming more bit depth widens the colour gamut, when depth sets the number of steps and the colour space sets the range.
Questions people ask
Bit depth, answered
The follow-up questions people search for once they have the definition.
Is 16-bit better than 8-bit for product photos?
Better to edit in, not to deliver. Work in 16 bits so tonal corrections have levels to spend, then export 8-bit JPEG or PNG for upload. The finished image looks identical at 8 bits and the file is much smaller.
Does higher bit depth mean more colours on screen?
Only if the display can show them. Most consumer screens are 8-bit or 8-bit with dithering, so a 16-bit file is displayed at 8-bit anyway. The benefit is in the editing, where the extra levels stop gradients breaking into bands.
Why does my sky or backdrop have visible bands?
Almost always 8-bit editing. A gradient occupying a narrow slice of the range gets stretched by contrast or exposure work, and the surviving levels spread far enough apart to be visible as steps. Redo the corrections from a 16-bit or RAW original.
What bit depth do camera RAW files use?
Typically 12 or 14 bits per channel straight off the sensor, depending on the camera and its settings. That extra headroom over 8-bit JPEG is the main reason RAW survives heavy exposure and white balance correction without banding or amplified noise.
Same cluster
More from the glossary
Neighbouring entries, so every term in the set is one hop from every other.
BMP
BMP is Microsoft's raster image format, which stores each pixel's colour value directly with little or no compression.
Bicubic interpolation
Bicubic interpolation works out each new pixel by fitting a smooth cubic curve through a 4 x 4 block of sixteen surrounding pixels.
Bokeh
Bokeh is the character of the out-of-focus parts of a photo, not simply how much blur there is.
From the studio
Tools that do this
Each one runs on your own device, so the photo stays where it is.
Sources
Where the facts on this page come from. Every link opens the specification, standard or documentation the claim was read at.
Keep reading
Elsewhere on Photomop
Platform image rules
Image size, file type and background rules for 22 selling platforms, each one sourced.
Read itConvert image
JPG, PNG, WebP and AVIF, with an honest answer about HEIC and alpha.
Read itAll the tools
The whole toolbox in three families, with what each one needs before it can run.
Read itOn-device privacy
The pipeline in five steps, what leaves your device, and how to check it yourself.
Read it
Now do it to a photo
Photomop is a photo studio that runs on your own device. Resize, crop, compress, convert, batch edit, remove a background and change one all work in the browser tab you are reading this in, at full resolution and with no watermark. The on-device page shows you how to check that the photo stays put.