Do megapixels actually matter?

The number counts area, not detail.

A megapixel is nothing more exotic than the pixel count of the frame, written in shorthand. A picture that measures 6000 across by 4000 down works out to 24 megapixels. That is the whole definition.

The part that matters, and that almost nobody is told, is that the count measures area while the detail you can actually see is set by the width and the height. That is why the relationship between the headline number and the visible improvement is not the one people assume. To make a picture twice as wide in pixels you need four times as many of them. A camera offering twice the megapixels of another is not giving you twice the detail; the gain along each edge is nearer to two fifths, which is a real difference and a far quieter one than the number suggests.

Every argument on this page follows from that. It is why the step from a small count to a moderate one is obvious, and the step from a large count to a much larger one is something you have to go looking for.

How little a portrait actually asks for.

Work backwards from the destination and the numbers turn out to be small.

• A profile picture asks for a fraction of one megapixel, because it is only ever displayed a few hundred pixels across.
• An 8 by 10 held in the hand, at the usual 300 dpi convention, works out to roughly 7 megapixels.
• A 16 by 20 read from a few feet back rather than under a magnifier sits comfortably below the density a held print asks for.
• A slide filling a 16:9 frame comes in near 2 megapixels.

Set that against what an ordinary modern camera produces and the gap is enormous. The pixels are not the constraint and have not been for a long time. If you want the arithmetic in full, including how to work out the largest sensible print from any file, how much resolution a use needs takes it apart properly.

What is actually doing the work.

Put two photographs of the same person side by side, one that looks expensive and one that does not, and the difference will not be pixel count. It will be some combination of a short list.

Light comes first, by a distance. Where it is placed, how large it is relative to the face, and how it falls across the nose and jaw decide whether features look sculpted or flattened. A big soft source close by does something entirely different from a small hard one further away, and no sensor has an opinion about it either way.

Then lens and distance, which together decide the shape of a face in the frame. Standing close with a wide lens enlarges whatever is nearest the camera, which is usually a nose. Stepping back and using a longer focal length lets features sit in proportion. After that: focus landing precisely on the near eye, an expression that belongs to the person rather than to the camera, and a background chosen on purpose rather than by accident. Choosing between white, gray or black behind you visibly changes a portrait in a way that a sensor upgrade never will.

The whole set of levers, ranked, is the argument in what makes a photo look expensive, and megapixels do not appear high on the list.

What extra pixels honestly buy you.

This is not an argument that the number is worthless. It buys specific things, and they are worth naming precisely so you can tell whether you need them.

• Cropping room, so one frame can be re-shaped for more than one destination without the smallest version running short.
• Very large prints seen from close range, where the viewer's nose is a foot from the paper rather than across a room.
• Fine retouching precision, because working by hand on skin is easier when the underlying detail is genuinely resolved rather than guessed at.
• Recovery from a frame that had to be shot looser than ideal, which is a rescue rather than a plan.

Notice that all four are about flexibility after the fact. None of them makes the photograph itself better at the moment it is taken. That is the honest position: more pixels widen your options later and change nothing about what is in front of the lens.

And what they quietly cost.

There are trades on the other side, which is why cameras are not simply built with ever larger numbers.

Packing more wells into the same sensor area makes each one smaller, and a smaller well collects less light, so each individual pixel is noisier. This is quieter than it sounds, because the sensor as a whole still gathers the same light, and viewed at one size the two files land close together. Where it does show is if you go looking at the picture pixel by pixel in dim conditions, which is exactly where the finest detail you paid for gets buried.

Very dense sensors also reach the limits of optics sooner. Stop a lens down far enough and diffraction begins to soften the image regardless of how many pixels are waiting behind it, so past a certain point the glass, not the sensor, is setting the limit on detail.

And then there is the honest human cost. More resolution renders skin more literally, which means retouching by hand takes longer and demands more restraint, since the goal is a person who looks like themselves rather than a surface that has been sanded. Files get heavier to store, slower to move around, and more of a nuisance to hand to anybody. None of that is a reason to avoid a capable camera. It is a reason to stop treating the number as a score.

Why the spec sheet misleads you.

A pocket device can advertise a pixel count that matches or beats a professional camera, and the pictures still do not compare. Nothing dishonest is happening; the number simply describes one variable while several larger ones sit outside it.

• Sensor area. Two frames with the same pixel count on very different sized sensors collect very different amounts of light, and light is the raw material.
• The lens in front of it, its size, its quality and how wide it can open, which decides how much of the scene arrives sharp and how the background falls away.
• The distance the picture was taken from, which is what draws a face flattering or unflattering before anything else has a say.
• Heavy automatic rendering, which smooths, sharpens and brightens on the way to the file and takes the decision out of anyone's hands.

The upshot is that the pixel count is the least informative thing on the sheet. Where the real gaps sit is a longer conversation, taken up in why a professional photograph beats a phone one, and it is mostly a conversation about light and control rather than about hardware.

For a working headshot, the practical answer is that the camera stopped being the limiting factor a long time ago. What limits the picture now is whether somebody is directing the light, the angle and your expression, and then finishing the file with care. If you want that whole process laid out from the beginning, the complete guide to professional headshots walks through it.