Basically every midrange or budget smartphone has atleast a 48 MP camera, while most of the flagships have about 13 MP camera. Then why are photos from the flagships better? Is it because of image processing or quality of lens or something else? And which specifications to look out for when buying a low cost phone?
In: Technology
The pictures you see from the Mars rover were taken with a 1MP camera. It just is behind really good glass.
Every sensor is a compromise. More pixels increase resolution but often at the cost of quality, particularly in low light. Fewer, larger pixels improve quality but reduce resolution and your ability to crop an image. You choose the sensor that suits what you shoot.
48 MP is only possible by making the sensors larger, meaning a larger phone, or by shrinking the pixels, which decreases their effectiveness. They are trying to overcome this, but it is still a work in progress.
Its important to remember that a resolution is simply a measure of size, not quality. It has more potential detail, but is not guaranteed.
Phones don’t actually have good cameras… well they do in a sense. What the cameras lack in quality and properties to capture the image well due to physical limitations of the camera. As in, no matter hoe expensive your phone is, it can not change the physics of light. So to get around this the phone manufacturers invest lot of time and effort to post processing the images. GPUs came to phones so they could be used to process camera images. All it does is try to compensate for the lack of physical objective and apparatus. This is why all sorts of “AI” which are actually just algorithms are used to basically approximate the actual picture.
Big DSLRs and other big cameras can focus on the actual storing of the information the sensor receives.
Basically phone’s camera’s pixels are poor quality and done with approximation, while biggers cameras have higher quality pixels which represent the informat what the sensor received.
There is a difference between, “this was probably the information I hot” and “this was the information I got”.
Because the number of dots you read is one tiny, tiny statistic about how a camera operates.
Lens quality and design, quality of each of those pixel sensors, the sensor range and sensitivity to each of the colours, post-processing performed on the camera chip and on the device itself, distances between lens and sensor and can the lens adjust/move itself and how far?, image stabilisation, …
It’s like saying “My car has 47 exhausts, why isn’t it faster than a Ferrari with only 1?”
It’s largely image processing.
Google’s pixel line has used the same camera sensor in it for the main camera since the Pixel 1.
With very good processing algorithms, you can get very good photos.
But these algorithms are very processor intensive, so it’s difficult for lower end devices to process photos using the same techniques, unless you want to wait a few seconds or more before being able to view the photo.
I do freelance photography, among other things, and have been doing it for years. I own a proper professional camera, understand photography principles, camera specifications, etc.
I preface my comment by saying this because people can often give answers which aren’t really based on in depth knowledge, and I want to establish that the written below is in fact going to be well informed.
Let’s first assume, for simplicity’s sake, that two different theoretical phones have the same sensor size, same lens quality, same focus mechanism quality, same ISO sensitivity (light sensitivity), same image processing hardware and software quality, etc – let’s look strictly at MP count. Let’s compare 12 MP with 48 MP.
Sensors are physical objects, which have limited real estate to contain pixels. Two equally sized sensors with different MP counts will have a difference in how large each pixel is physically on the sensor. In our case, a 12 MP sensor will have larger pixels than a 48 MP sensor.
Larger pixels means that each pixel captures more light individually. Smaller pixels thus means that each pixel captures less light individually. This single simple difference means that when comparing two sensors that are otherwise exactly the same, except for their pixel count, the sensor with 12 MP will have a better light sensitivity than the 48 MP sensor.
Because we also assume the same exact hardware and software, this means that the smaller pixels of a 48 MP sensor don’t get the additional special treatment that they would normally need to capture light as well as the 12 MP sensor does.
So, point 1: A 12 MP camera has bigger pixels than a 48 MP camera, letting them capture light better. The bigger pixels thus produce a better dynamic range, which in simplified terms means that the range and depth of color captured by the 12 MP camera will be better. The bigger pixels will also capture more detail from darker areas, because they are more sensitive to light and to subtler light variations. Lastly, the bigger pixels will operate better under lower light conditions, because, again, they capture more light.
Point 2: Larger pixels also means that they are less sensitive to tiny variations in sharpness. Because the lenses and focus mechanism on both of our theoretical phones are the same, this means that both sensors get the exact same image projected onto them. Because there is no lens that is perfect, and no focus mechanism that can focus “perfectly”, there is always going to be some level of blurriness even at the sharpest edges. A 12 MP sensor will be much less sensitive to this blurriness, producing a picture that looks sharper than the 48 MP sensor. The 48 MP sensor will produce a picture where tinier errors in focus are a lot more obvious and visible.
Point 3: There is a term called cross-talk, which in this context refers to electrical signals bleeding from one pixel to its neighbors. Each pixel converts light that hits it into an electrical signal. The electrical signals generated by each pixel can bleed to it’s neighbors because of the physical limitations of building physical circuitry – it is never perfect, because there is no such thing as perfection. Thus, the smaller and denser the circuitry, the higher the chance of cross-talk between neighboring pixels. Our 12 MP camera has a less dense circuitry than our 48 MP camera, and thus less cross-talk. There is also less noise in our 12 MP camera, for similar reasons.
Point 4: Because our theoretical phones are equal in all respects except for MP count, they are also equal in their image processing capabilities. This one is simple to explain – a 12 MP image is faster to process than a 48 MP image.
Point 5: This one is less about the sensors themselves, and more about the manner in which sensor technology is being innovated over time. Simply put – a 12 MP sensor is likely a well established sensor, one which has existed for longer. Engineers had more time to learn how to work with this sensor, how to build hardware in the best way for this sensor, how to fine-tune software in the best way for this sensor, and they had more time to develop new versions of this sensor which are more refined, while leaving the fundamentals of it intact.
In reality, our two phones will likely not be the same. This means that the 48 MP camera phone will need to ultimately be more expensive than the 12 MP camera phone in order to achieve the same exact light sensitivity, same exact processing speed, same exact cross-talk resistance, same exact sharpness (by investing more into the focusing mechanism, the software controlling the focus mechanism, and the lens), and the same exact processing quality as the 12 MP camera phone.
When looking at a phone that offers 48 MP, but is cheaper than one which offers 12 MP, it is not hard to assume that these 48 MP are achieved by various compositing trickery, and that ultimately the quality produced by this 48 MP camera will be worse than the quality produced by the more expensive 12 MP camera.
Latest Answers