photomopOpen the studio

Resolution and scaling

Pixel

Also called px, picture element or pixels. Here is what it means, when it changes what you export, and what people get wrong.

Resolution and scalingRuns on your deviceUpdated 10 August 2026

The short answer

A pixel is the smallest single dot of colour an image is made of. A digital photo is a rectangular grid of them, each storing one colour value, usually 8 bits each for red, green and blue. A pixel has no physical size until you print or display it.

Updated

The short version

Pixel at a glance

Quick facts about Pixel: short form, typical depth, colours from 8-bit rgb, physical size, shape in a photo
Quick factDetail
Short formpx
Typical depth8 bits per channel, 256 levels each
Colours from 8-bit RGB16,777,216
Physical sizeNone until an output size is chosen
Shape in a photoSquare sample on a regular grid

Slide the table sideways to see every column.

How much information does a single pixel hold?

In a standard 8-bit photo each pixel stores three numbers, one each for red, green and blue, and each number runs from 0 to 255. That is 256 levels per channel and 16,777,216 possible colours per pixel. Formats that carry transparency add a fourth number, alpha, for opacity. Editing formats such as TIFF and camera raw often use 16 bits per channel, which leaves more room to push exposure and colour before banding shows up.

Why does a pixel have no physical size?

A pixel is a sample, not an object. The file stores a grid of values and says nothing about how wide each one should be. Size only appears at output. Print a 2000 pixel wide image across 10 inches and each pixel occupies 1/200 of an inch; print the same file across 5 inches and each pixel is half that. On a high-density screen one image pixel is often drawn across two or three hardware pixels.

Is an image pixel the same as a screen pixel?

No. An image pixel is one sample in a file. A screen pixel is a physical cell in a panel, normally built from three subpixels, one red, one green and one blue, sitting side by side. Text rendering sometimes drives those subpixels individually to fake extra horizontal resolution, which is why a screenshot of text can show coloured fringes when you zoom in. Those fringes live in the capture, not in the original image.

What people get wrong about pixel

Each one is a real failure mode, not a style preference.

  1. Assuming a pixel has a fixed size in millimetres, when its physical size only exists once you pick a print or display size.

  2. Confusing an image pixel with a screen subpixel, the individual red, green or blue element inside a display cell.

  3. Expecting a pixel to hold part of two colours; it stores exactly one value, which is why diagonal edges look stepped.

Questions people ask

Pixel, answered

The follow-up questions people search for once they have the definition.

How many pixels are in a 4000 x 3000 photo?

Twelve million. Multiply width by height: 4000 x 3000 = 12,000,000 pixels, which is 12 megapixels. Every one of those pixels stores its own colour value, so pixel count is the main driver of how much raw data an image contains before compression.

Is a pixel always square?

In photographs and web images, yes. Some legacy video formats used non-square pixel aspect ratios, where each sample was stored wider or taller than it was displayed, but every common still image format today assumes square pixels on a regular grid.

Can you have half a pixel?

Not in a stored file. Pixel counts are always whole numbers. When you resize or rotate, the software computes fresh whole pixels at new positions from the old ones, which is why a resize is never a perfectly reversible operation.

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.