Is This Photo Good Enough to Print?

The question is almost always asked as “how many megapixels do I need”, and megapixels are the part you are least likely to get wrong.

The resolution part, quickly

The old rule is 300 DPI, and it is right for something held in the hand. It is wrong, expensively, for anything else — because the real variable is viewing distance.

A billboard is printed at around 15 DPI. Nobody stands close enough to notice.

So a 12-megapixel phone photo prints beautifully at A3. If a print looks bad, resolution is rarely why — which is unfortunate, because resolution is the one number everybody checks.

The four things that actually decide it

Focus. Print is unforgiving in a way screens are not: a screen shows you a fraction of the pixels, a print shows you all of them at once, at whatever size you chose. Softness that was invisible on a laptop is the first thing you see on a wall. This is the number one cause of a disappointing print, and it is entirely checkable in advance.

Noise. High-ISO grain that reads as texture on a backlit screen reads as dirt on paper, because paper has no backlight and much less dynamic range. Ink also spreads slightly, which merges fine noise into mush rather than resolving it.

Blown highlights and crushed shadows. A screen renders pure white as light. Paper renders it as nothing — the bare paper. Any detail clipped to white is not dim in a print, it is absent. Shadows are worse: the darkest black most printers achieve is far lighter than a screen’s black, so everything near the bottom of the range compresses together.

How many times it has been saved as a JPEG. Every save re-quantises. A photo edited, saved, reopened, edited and saved again has been through the mill repeatedly, and the artefacts are cumulative and permanent. The giveaway is blocking on smooth areas — a sky, a wall, skin.

Checking the JPEG part without guessing

There is a genuinely useful trick here. A JPEG’s quantisation tables are the standard table scaled by the quality setting the encoder used — so the setting can be read back out of the file, exactly, with no decoding and no guesswork.

Anything at quality 90 or above is fine for print. Below about 75, expect visible blocking on smooth areas at size. It is worth checking before you find out at A2.

The pre-print checklist

1. Focus at 100%, on the part of the frame that matters. Not the whole frame — a portrait at f/1.4 is meant to be soft nearly everywhere.
2. Clipping. A few blown specular highlights are normal and fine. Large flat areas of pure white or pure black are detail that will not exist on paper.
3. JPEG quality, if it has been through a JPEG stage at all.
4. Noise, judged at the size you will print, not at 100% on a monitor.
5. Resolution last, against the table above rather than against 300 DPI.
6. Soft proof in the print profile if your software does it. Colours outside the printer’s gamut — saturated blues and greens especially — shift, and it is better to know.

Check a photo

Upload one image and DiffALL reports focus, noise, compression and exposure separately, each with its own verdict and the measurement behind it, and it marks where on the image it found the problem. Where the file is a JPEG it reads back the quality setting it was saved at; where the EXIF is intact it reports the ISO beside a noise finding and the shutter speed beside a soft one — because “soft” and “soft at 1/15 of a second handheld” are different diagnoses.

There is no single overall score, on purpose. A photo that is perfectly exposed, clean, 24 megapixels and out of focus is not 75% good.