Samsung and Omnivision Claim Smallest Camera Pixels

> AFAIK you want the biggest pixels you can for the resolution you want in order to cut down on noise, no?

You want the biggest sensor, but nowadays, we're targeting smallest pixels at the same time.

The basic idea is that every pixel will have its own quantum and thermal noise, the noise on every pixel is independent. This means that to have the best measurement of the "real" light brightness at one place, you make the median of all the surrounding pixels (that's called binning). This obviously reduces resolution, hence we're making smaller pixels. (Nowadays we make better algorithms than median, but median is still pretty good)

Lately, one of the major reasons we've been losing light on the sensors, is the RG/RB matrices, we loose light at every transition from one color to another.

So, you combine those two pieces of information, about pixel binning and about color filtering, and that makes you modern sensors: 108MP (in Samsung Galaxy S22 Ultra case), where the RG/RB matrix only has 12M"spots", pixels being 9-binned (so you "create" one pixel, by measuring 9 at the same time).

So Galaxy S22 Ultra's 108MP is kinda "just" 12MP with much better light sensitivity.

In real life, it's a bit better than just 12MP, because even if you are in a R-only region, you can still find the changes of global brightness, and have sharpness further than 12MP. Hence Samsung touts "combining 9-binned image with non-binned image"

A lot of modern flagship Android phones use pixel binning. The Galaxy S21 Ultra has a 108MP camera to take 12MP photos for example. Sensor size still matters and part of why those Nokias were so good at the time is they just had significantly bigger sensors than most phones.

>Fitting more and more smaller pixels on the same sensor doesn't improve anything at all (lower DR, lower low light perf

They improve sharpness if optics can resolve them. Also, having more data will might ease autofocus algorithms and computational photography techniques.

Low light and DR depend solely on the quantity of light falling on the sensor. That depends on scene illumination, sensor size and lens aperture,not on pixel size.

However, having a huge resolution is a waste of storage and processing power if the optics can resolve much of those pixels. And it could slow things down.

Typical big numbers game from Samsung. As you said, it definitely sells.

Omnivision are not exactly known for image quality either. They're known more for being cheap and (mostly) decent for the price, but tend be noisier than e.g. Sony's sensors.

Yeah we need innovation in Lens. And some other breakthrough with sensors. But I dont see anything coming within the next 3 years though.

Yes, it means that i'll be diffraction-limited, even if the lens is F/1.0. It'll be diffraction-limited for red light around F/0.9. It's possible to make lenses that open, but hard. That's why camera phone lenses are so funky.

Why not if you can use software to reconstruct the original image?

Unless you have a telescope with an extremely wide aperture, then no this sensor doesn't help. Pixels that small mean it is diffraction-limited if the telescope has an aperture of F/1.0. That's plausible in the most funky of camera lenses (but expensive), but completely unheard of in telescopes. The widest aperture telescopes are the RASA-type which have an aperture of around F/2.0, and they are expensive enough. They are regarded as fantastic light-sucking monsters, great for dim targets but not for small targets.

Most of the time, the image sensor on a telescope actually slightly over-samples the sharpness of the image that the telescope is able to project. So having smaller pixels won't help.

But to actually answer your question, these small pixels will be worse at low-light applications than larger pixels.

You want larger pixels to collect more light for astronomy.