Differences in the colours of sensors?

hsphsp

Well-known member
Messages
167
Reaction score
192
Location
Oulu, FI
If I remember correctly, about 45 years ago, the experts generally said that Kodak was slightly better in face colour than the others and Agfa colours were slightly better in natural green than the others.

Question: are there similar differences in the sensors of today's digital cameras? I mean the raw file from the sensors/cameras with a neutral profile, without any further processing. There are probably differences in weighting between the JPG files available from cameras from different camera manufacturers.

Or whether such a question is even reasonable. As far as I know the Bayer filter is commonly used, while Fujifilm uses the X-Trans filter. So there are only two basic sensor technologies. And the sensors are nowadays CMOS based. Is that so?

This topic has been discussed a bit in the thread: : https://www.dpreview.com/forums/post/67678666 / Why do all "neutral" camera profiles look different?
 
The formulation of the color filter array varies, and the precise chemical composition of the dyes and their density will have a major uncorrectable effect on a sensor’s specific color response.

In particular, there will be some colors—observed in controlled lab conditions—which will be distinguished by the eye but not by the raw camera response, and vice versa. The general principle is called metamerism and metamerism failure:


Humans and cameras can’t individually detect every wavelength of light individually, but can only detect broad frequency bands, but the really critical fact is that the overall spectral sensitivity of these bands don’t match between humans and cameras, and the camera models usually don’t match other models, even within the same manufacturer. (Humans don’t match either, but outliers tend to be well-characterized, such as with the various types of color blindness.)
 
If I remember correctly, about 45 years ago, the experts generally said that Kodak was slightly better in face colour than the others and Agfa colours were slightly better in natural green than the others.

Question: are there similar differences in the sensors of today's digital cameras? I mean the raw file from the sensors/cameras with a neutral profile, without any further processing. There are probably differences in weighting between the JPG files available from cameras from different camera manufacturers.

Or whether such a question is even reasonable. As far as I know the Bayer filter is commonly used, while Fujifilm uses the X-Trans filter. So there are only two basic [color] sensor technologies.
Unbeknownst to some, there are two kinds of color sensor - one with a CFA and one with photosensitive layers buried in the silicon .(no filters).
And the sensors are nowadays CMOS based. Is that so?
Not all, depending on when "nowadays" is ...
This topic has been discussed a bit in the thread: : https://www.dpreview.com/forums/post/67678666 / Why do all "neutral" camera profiles look different?
Because "neutral" means all things to all men. :-)
 
If I remember correctly, about 45 years ago, the experts generally said that Kodak was slightly better in face colour than the others and Agfa colours were slightly better in natural green than the others.

Question: are there similar differences in the sensors of today's digital cameras? I mean the raw file from the sensors/cameras with a neutral profile, without any further processing. There are probably differences in weighting between the JPG files available from cameras from different camera manufacturers.

Or whether such a question is even reasonable. As far as I know the Bayer filter is commonly used, while Fujifilm uses the X-Trans filter. So there are only two basic sensor technologies. And the sensors are nowadays CMOS based. Is that so?

This topic has been discussed a bit in the thread: : https://www.dpreview.com/forums/post/67678666 / Why do all "neutral" camera profiles look different?
You may be looking in the wrong place.

 
The formulation of the color filter array varies, and the precise chemical composition of the dyes and their density will have a major uncorrectable effect on a sensor’s specific color response.
I think the dyes are made by Fuji.
In particular, there will be some colors—observed in controlled lab conditions—which will be distinguished by the eye but not by the raw camera response, and vice versa. The general principle is called metamerism and metamerism failure:

https://en.wikipedia.org/wiki/Metamerism_(color)

Humans and cameras can’t individually detect every wavelength of light individually, but can only detect broad frequency bands, but the really critical fact is that the overall spectral sensitivity of these bands don’t match between humans and cameras, and the camera models usually don’t match other models, even within the same manufacturer. (Humans don’t match either, but outliers tend to be well-characterized, such as with the various types of color blindness.)
 
The formulation of the color filter array varies, and the precise chemical composition of the dyes and their density will have a major uncorrectable effect on a sensor’s specific color response.
I think the dyes are made by Fuji.

They aren’t the only maker. Fujifilm uses inorganic pigments, which limit pixel size, as these are small particles suspended in a binder. Fujifilm reportedly has a wide variety of pigments available.

This company sells dye based CFA materials:


Dyes aren’t particularly durable, fading with light exposure, but they may be used with tiny pixels.

From what I understand, CFA application is tightly integrated with chip manufacturing on the same production line. It’s difficult to find any online information on what is specifically used with a particular camera model.
 

Keyboard shortcuts

Back
Top