I get that more MP is better ends up being pure marketing. However I don’t understand why all cameras don’t bring 108 MP sensors or whatever. I’m confused.
In: 16
Image quality comes down very heavily to optics, and dedicated cameras have far more powerful optics than phones. A major factor of optics is size. Cell phone lenses are nowhere near big enough to do what a dedicated camera lens can.
Smartphones pixel bin, so the image you actually get is 12 megapixel. In essence, the sensor ‘bins’ pixels into groups to create one larger pixel out of nine smaller ones.
Professional photographers typically don’t need higher than 20 megapixels for any reasonable print or screen size. What is more important is the quality of each of those photosites, some are known to be excellent. Sony makes some outstanding 24 megapixel sensors that are much larger than a smartphone. Sony makes most of the sensors (by now, all?) for Nikon cameras. At any rate, when you smear those pixels over a larger area, you end up with much better low light capabilities in most instances. People still shoot cameras from 2008 with 12.4 megapixels on a full frame (so 35 mm, same size as standard film) sensor that is now older than some people reading this comment.
We are now seeing what are called ‘backlit’ CMOS sensors and ‘stacked’ sensors which have comparably few pixels (I think Nikon’s biggest is 46.7) but those technologies allow for faster autofocus and image detect autofocus. A Z9 can literally tell what a bird is, when it is in focus, and can take pictures of it automatically. They have car recognition and eye recognition, all of that built off of the sensor’s electronics so they can go as fast as an old school DSLR but have a lot of the high tech features of a smartphone. The MP count matters less than the overall quality of the sensor and the sensor’s capabilities.
In reality, when you view the image, your display device is less than 2MP. As a result, any imaging device above 2MP is meaningless.
Here’s a weird analogy. Taking a picture is like ushering millions of tiny points of light, all different colors and brightnesses, into an enclosed space (say, a sports arena) through a single door and then making them sit down in the right order, and then quickly recording where they all sat before they disappear. Then your recording is a picture, and it looks more or less like the scene outside your stadium door.
The seats are pixels in your sensor.
The lens is like a group of ushers that tell each piece of light where to sit. When light comes in, they point in a straight line to where each photon should sit, and the photon changes course toward its seat. The lens isn’t perfect, though, so sometimes the photons go a little too far off track, or clump together around the edges of the stadium, etc. because the angle they “should” go is hard to distinguish.
Now, to improve the quality of your pictures, you have some options. One is to wait longer so multiple photons come and go, and average where they sat instead of taking one sample (long exposure). This will get blurry though if the scene changes at all.
One is to increase the number of seats. That way there are more possibilities of where a photon can sit, meaning more resolution. However, if you do this without making the stadium bigger, the lens’s job is much harder now. Even the slightest error in where it points will send the photon to the wrong seat, causing distortion.
If this happens, you could write some software that is good at catching errors in the record and applying corrections. (Apple is amazing at this).
Alternatively, you could make the stadium bigger, and make each seat bigger. Now the lens has a much easier job pointing each photon to exactly the right seat, even with the same number of pixels.
With a bigger stadium, you can also let more photons in at a time, meaning a larger sample even with a short exposure, and also fit a much bigger lens, which will be more precise (the analogy kinda breaks down here).
So they fact that we can capture a 100MP image from hardware that fits in our pocket is an amazing feat of engineering. But it does a bit of a different job than a full-blown camera.
Latest Answers