- Phone camera resolution: 64 megapixels, 2020
- Current camera (Nikon D80 DSLR) resolution: 10.2 megapixels, 2006
- Next camera (Nikon D7000 DSLR) resolution: 16.2 megapixels, 2010
- Highest-res Nikon camera I’ve found (Z7, mirrorless): 45.7 Megapixels, 2018 (ish?)
That’s a hmmm from me. But don’t get me wrong, just because the resolution is higher, does not mean that it’s a better camera. There’s so much that actually goes into a camera that just a single number isn’t telling you. Guess that means that today, we’re taking a deep-dive into camera construction, aren’t we?
Parts of a Camera
Okay, let’s start with some very quick basics: Every camera starts with a lens, which is used to focus the light onto the sensor. Different lenses can do different things, some lenses, like tilt/shift lenses, can do some really cool things, but there is no such thing as a lens-less camera.
Some cameras (D/SLRs) have a physical shutter that is used to expose the sensor, others (phones, mirrorless, etc.) don’t, and use an electronic shutter instead, where the sensor itself is starting and stopping recording, and there are no physical parts to do the work.
Also, after the sensor output, something has to decode the color filter pattern, interpret the raw returns and make a full image. This is the job of the image processor, a bit of software / firmware that takes raw input on one end, and spits out 100% standard images on the other.
Finally, the actual sensor is an important part too, for one reason: size.
Why MP Aren’t Everything
First, have you noticed how compact the entire camera package is on your phone? I’ve seen SLRs (for the purposes of this post, SLR refers to both conventional SLRs and DSLRs) with lenses the size of someone’s entire arm, and I doubt manufacturers are making their products bulky and unwieldy for no reason. The lens in a mobile phone camera just cannot be compared to a proper SLR lens. Many of the features of an SLR like aperture adjust, optical zoom, in-lens vibration reduction, and a multitude of other things cannot be done in that short package (Apple is making 2x optical zoom cameras though).
A lack of aperture adjust means no way to change the depth of your depth-of-field, and a lack of optical zoom relies on digital zoom, the equivalent of cropping a picture down, then enlarging it, to make things look bigger. You do so at the cost of throwing away a lot of data since you can’t add data that wasn’t originally captured, and digital zoom is a post-process effect (or, an image processing effect). Taking a 20 megapixel picture at 4x optical zoom means you’re actually taking a 5 megapixel picture in terms of actual details.
Phones do not have a mechanical shutter, period. There’s just no space. Admittedly, any SLR with a live view option, or any mirrorless, also have digital shutters, so there’s that. Really both have their ups and downs. I can’t say that one is objectively better than the other, but it’s still a difference.
An SLR has one job: to take pictures. To that end, all the firmware in the camera should (hopefully) be written for the express purpose of that one job. Cameras cost more than phones, but do less, therefore, we expect that while they only do one job, they will do that job to a much higher degree of quality. Phones, on the other hand, at this point, are just pocket-computers. Their job is to do, well, pretty much anything, really. It’s unreasonable to expect that a $900 device that can do pretty much anything you ask will be able to compete with a device that could potentially have had just as much money spent just optimizing the programming. SLR image processing has, and always will be, superior to the processing algorithms found on phone cameras.
Plus, they also have one other benefit: Just about any good DLSR will let you save raw files, the exact format will vary from manufacturer to manufacturer, I’m a Nikon person, so for me, that means NEF files. Raw format is about as close to the sensor that you can get, with as little image processing as you can get. I can then take these raw files, and load them onto a much more powerful device, that can take much more time processing, and allow me to tweak each of the values in the processing pipeline myself to get the shot to come out (almost) exactly how I wanted it to. While you can post-process a processed JPEG file, it’s nowhere near the amount of flexibility that raw files can give you.
Finally, sensors. In the end, a CMOS chip is a CMOS chip, but the size of that chip is important. A full-frame 35mm sensor is the actual size of a 35mm film frame, so about 36 by 24 mm. Compare that to the roughly 7 by 6 mm sensor in my phone, one is a lot larger. Really, the bigger the sensor, the more actual information it can take in within the sensor’s resolution limits. Smaller sensors also have a higher theoretical maximum resolution, at which point, digital upscaling is used to produce the final image.. essentially just zooming in. Again, you can’t actually add detail onto where there is none to begin with.
While having a higher resolution photo (or video) is nice to work with, as it means that it feels like I have more freedom to crop, zoom, and edit to fit my needs, just because one camera has a higher megapixel output does not in any real way indicate it’s quality, there’s too many other factors at play to make that call, and I didn’t even name all of them, just the most important ones!
My phone may beat Nikon at the final image resolution. Nikon beats my phone at the final image quality.