Image resolution explained

Most arguments about resolution are two people using the word for different quantities. There are three, and they are related but not interchangeable:

• Pixel dimensions — width × height in pixels, e.g. 4000 × 3000. This is the amount of information in the image, and the only one of the three that is genuinely fixed.
• Pixel density — PPI (pixels per inch), or DPI as it is loosely called. A ratio: how many pixels are packed into an inch when the image is printed or displayed at a given size.
• Physical size — how large the image is in inches or centimetres once it exists in the world.

The image file stores the pixels. It may also store a density tag — JPG in its JFIF header, PNG in a pHYs chunk — which is a note about intended print size, not a property of the pixels.

Change that tag and no pixel moves. A 3000 × 2000 photo tagged 300 PPI and the same photo tagged 72 PPI are byte-for-byte identical in image content; they only differ in what a print dialog suggests as a default size. That is what the DPI converter writes, and what 300 DPI means goes into it further.

Some formats have no density field at all — WebP is one — which is a practical reason to deliver print files as JPG or PNG rather than WebP.

On a screen, the density tag is ignored entirely. A browser lays out an image by its pixel dimensions and the CSS you give it; nothing reads the PPI field. This is why “save for web at 72 DPI” is a habit left over from 1990s monitors rather than a requirement.

What does matter on screen is the device pixel ratio. A phone advertising a 400px-wide viewport may have 800 or 1200 real pixels across it, so an image displayed 400px wide will look soft unless the file is roughly two to three times that in real pixels. The practical rule for web work is to supply about twice the CSS display width, then compress — covered in optimising images for websites.

Print is where density becomes real, because ink lands on a physical surface. The common targets:

Multiply the finished size in inches by the density you need. That is the pixel count required.

• A 6 × 4 inch print at 300 PPI needs 1800 × 1200 pixels.
• An A4 page (roughly 8.3 × 11.7 in) at 300 PPI needs about 2480 × 3508.
• A 24 × 36 inch poster at 150 PPI needs 3600 × 5400.

If your file has fewer pixels than that, you have three honest options: print smaller, accept lower density, or reshoot. Check the pixel dimensions first — every tool here shows them on upload, and the resizer shows them before and after.

Enlarging an image invents pixels by interpolating between the ones you have. Interpolation can be smooth and it can be clever, but it cannot recover detail that was never recorded — it produces a larger, softer version of the same information.

What that means practically

• Upscaling to satisfy a stated minimum resolution rarely passes review. Stock agencies in particular are good at spotting it, because interpolated detail has a characteristic softness.
• Modest enlargement — say 10–20% — is usually unremarkable. Doubling is visible on anything with fine texture.
• Downscaling is the opposite case and is genuinely lossless-looking: you are discarding information you had, and the result is sharp. Doing it well is mostly about not doing it in the wrong order.