WAKE UP: YOUR SMARTPHONE CAMERA IS NOT BETTER THAN A 15-YEAR-OLD CAMERA

You know what, guys? I am really tired of reading that today's smartphones produce better images than cameras from ten or fifteen years ago. They don't.

The argument almost always starts with megapixels. “Look, this phone has 200 megapixels.”

Okay. And? Megapixels are not a measurement of photographic quality. They tell you how many pixels the manufacturer is putting into the file. They do not tell you how much real detail the optical system has actually resolved, how natural that detail is, how the tonal transitions behave, or what happens when you start looking closely at the image. This is where I think we need to wake up and talk about physics.

PHYSICS HAS NOT CHANGED AND WILL NOT CHANGE IN THIS PLANET

A smartphone has a tiny sensor and a tiny optical system. Even the most expensive phones use sensors dramatically smaller than those found in conventional cameras. That has consequences for light gathering, noise, dynamic range and the ability to resolve and preserve fine detail. And then there's the lens. A lens determines how much actual information reaches the sensor. You can put 50, 100 or 200 million pixels behind a tiny lens, but that doesn't mean the lens can resolve 200 million pixels worth of genuine photographic detail. More pixels are not automatically more information. This shouldn't be controversial. Yet somehow it has become one of the most persistent myths in photography.

THE IMAGE IS NO LONGER JUST A PHOTOGRAPH

Then comes computational photography. Modern smartphones don't simply record an image. They analyse it, merge exposures, suppress noise, sharpen edges, reconstruct details and manipulate local contrast before you ever see the final result. And to be fair, the results can be spectacular. On a phone screen. That's the trick. And the big misunderstanding. Taking advantage that more and more people don’t use a computer anymore. Because at normal viewing size, computational photography can produce an image that looks remarkably clean, sharp and detailed. But try to open the original file on a computer and start looking closely, however, and another picture emerges. Faces can become smeared. Hair can turn into a strange painted texture. Fine detail can look like a pastel drawing. Textures become waxy or watercolor-like. Edges acquire artificial sharpness. And details that were never actually resolved by the optical system may have been reconstructed by software. All thjat to me looks like a huge pile of crap. At 100%, sometimes even at 200%, you can see the difference between information that was captured and information that was manufactured.

CLEANER DOES NOT MEAN MORE DETAILED

This is particularly obvious with noise reduction. Small sensors produce more noise under many shooting conditions, so smartphones rely heavily on computational noise reduction. But noise and fine detail are not completely separate things. Remove too much noise and you remove texture with it. The result is cleaner, but not necessarily more informative. This is why a smartphone photograph can look incredibly polished while having surprisingly little genuine microdetail. And digital zoom or aggressive cropping only makes the problem more obvious. There is simply no algorithm that can turn missing optical information into genuine captured detail. But it cannot change what the lens and sensor physically recorded.

THIS IS WHY I KEEP TO WORK WITH MY 5D MARK II

My Canon 5D Mark II was released in 2008. It has 21 megapixels. Modern smartphones can have several times that. And yet when I open the files and actually examine them, the difference is obvious.

The Canon file has a photographic structure that feels natural.Real detail.Natural texture.Natural tonal transitions. Noise and grain that belong to the imaging process rather than being aggressively removed and reconstructed. It isn't a perfect file. It isn't particularly modern either. But it is a photographic file. A real, organic photographic file. A smartphone can produce something that looks more impressive at first glance. That doesn't mean it contains more real photographic information. That is where is the big scam with computational photography.

THIS IS WHERE PHOTOGRAPHERS SHOULD CARE

I completely understand that most people don't need any of this. Maria Calzetta doesn't give a shit about what I'm writing here. If she wants to photograph her dog, her lunch or her holiday and post the result on Instagram, that's perfectly fine. A smartphone is extraordinarily good at that. The problem starts when photographers claim that smartphones have therefore replaced cameras because they produce “better images.” Better for what? Are really photographers those saying this bullshit? That's the question.

Excuse me, I work editorially. My standards are different from those making photos just for the instagram. If I open a file at 100% and the face is already smeared, the hair has become some kind of crappy pastel texture, fine details have been reconstructed and shitty artifacts are appearing everywhere, I don't consider that a professional file. Not even a photograph. Especially when that photograph has to sit alongside images made with a proper camera. At that point, the argument about how beautiful the image looks on a phone becomes irrelevant. The file is what matters.

LOOK AT THE FILE

This is the simple test. Take one of those supposedly amazing smartphone photographs. Open the original file in Lightroom. Put it on a proper monitor. I mean…a real monitor, not the fuckin’ smartphone.

Now Zoom to 100%. Look at the face. Look at the hair. Look at skin texture. Look at distant objects. Look at fine patterns and edges. Don't judge the photograph by the tiny screen it was designed to impress. Inspect the file. Yes, now you can throw your phone out the window hahahaha. Because that's when you begin to see what the camera actually captured and what the software had to invent, reconstruct, smooth or destroy. And this is the distinction I think we have completely lost: a convincing image is not necessarily a high-quality photographic file. If your scope and goal are the same as your aunt's, or those of people who exist only on Instagram, fine, keep shooting with that crap. If you are a photographer, I am sorry but the smartphone is not something to consider at all.

WAKE UP

Smartphones have become extraordinary imaging devices. They are small, convenient, connected and incredibly sophisticated. Their computational photography can produce images that are astonishingly good for everyday use, for those that are not interested to learn how to manage a file is perfect.

But none of that changes physics. A tiny sensor doesn't become a large sensor because the marketing department writes “200 MP.” A tiny lens doesn't become a professional optical system because software reconstructs the missing detail. And an algorithm cannot repeal the limitations of the imaging system. As I said open the file on a computer and look at it properly. Sometimes Lightroom gives you a very simple reality check: **The photograph looked great on your phone. The file is another story.**

Wake up.

Wake up, guys, because camera manufacturers are going everyday towards the computational photography. And that is affecting already the results with some new mirrorless. I didn’t understand even the fact the Leica Q presented the possibility to make zoom with the fixed lens of that camera: it is a no sense because any real photographer knows that is a digital zoom that is a crop in function, a digital zoom.

Or look how conservative are now designed the new mirrorless cameras. They already present many features that are good for those that are not real photographers.

Computational Photography needs to go…outside of any serious discussion about who the photographer truly is.

IN CONCLUSION

I know this piece is going to spoil the party for a lot of people. Don't take it too hard. I know, it’s tough. Especially if you spent a fortune on a smartphone just for the camera. That is why I urge you to stop listening to the siren calls of marketing and start thinking for yourselves. Rely solely on direct experience. Conduct your own tests: don't let others do it for you.