Knowing there are different sized sensors and a nearly unlimited combination of lenses with the various cameras, since RAW captures RAW light data, how much does the sensor itself play a role in making up the final image since there is no color information (just monochrome, because color is how our eye perceives the image) - correct me if I'm wrong there...

Is it more the lens and how the camera processes the image from the sensor than the sensor itself? What makes one sensor better than another if shooting in RAW?

  • :| what do you mean by "there is no color information"? the typical sensor captures each primary colour in different pixels
    – fortran
    Jul 2, 2013 at 23:34

3 Answers 3


The sensor have a huge affect on the image quality. Many parameters that related to the sensor will affect you RAW data. I will only give some basic info:

  1. Different sensors have different pixel size: Bigger pixel increase the dynamic range and reduce noise - 2 key factors in image quality.
  2. The shape of the "pixel": Some sensors have more advance design that enable using more of the light coming through the lens.
  3. The sensor technology:


    You can see that the Foveon sensor structure is totally different, hence, produce a different RAW data.

The lens does not have a role in the processing. The processor can get data about the lens, for example, the lens shading or the lens chromatic aberration and use this data to process the image differently, but this does not relate to the RAW data.

  • 2
    As much as the marketers would have you believe otherwise, the Fovean chip does not collect all of the light. There is light loss as the the light passes through each layer of the chip. The blue layer absorbs some of the green and red light, the green layer also absorbs some of the red light.
    – Michael C
    Jun 27, 2013 at 16:00
  • That's true, but it does not change the fact that the RAW data is represented differently.
    – Itay Gal
    Jun 27, 2013 at 16:03
  • 2
    It is different, but has its own set of advantages and disadvantages compared to monochromatic sensors using a Bayer mask. The direction of the future seems to be in using a Bayer type mask with a non-repeating 6x6 pattern of 20 green and 9 each red and blue sensels the way Fuji's latest sensors are designed. This allows for the elimination of most fixed pattern noise without the need for an anti-aliasing filter. The advanced demosaicing algorithms made possible by the computing power of current microprocessors is also significantly driving the image quality possible with a Bayer type chip.
    – Michael C
    Jun 27, 2013 at 16:10
  • 1
    @MichaelClark - I did not notice Itay saying Foveon sensor is better, not even a hint at that direction if not counting the (rather poor) choise of image shown with the answer. Jun 27, 2013 at 19:53
  • @EsaPaulasto But that image is part of the answer, and by and large the part to which I was responding.
    – Michael C
    Jun 27, 2013 at 20:22

RAW does not capture raw light data. Rather, it records the digitized version of the analog electrical data produced by the image sensor as it collects light. The quality of that data is totally dependent upon the image sensor's sensitivity to light, the base signal produced by the sensor (noise), and the quality of the digital to analog conversion of the sensor's output.

In general, the primary factor that determines a sensor's sensitivity to light is the size of its sensels or photosites, often referred to as pixels (Pixels, properly used, refer to the output portion of an imaging sensor after the data has been demosaiced that corresponds to each sensel location on the sensor). This is so because the larger a sensel is, the more photons it captures for light of a certain intensity falling on the sensor. In order for a sensor to have a relatively high sensitivity to light and a high resolution output the sensor needs to be large enough for all of those larger sensels to fit on the sensor chip. The way the electrical signals produced by the sensor are processed also plays an important role in the equation, but the processing engine can only work with the information the sensor has collected.


It sounds like there is some general misconceptions from your question. When shooting RAW, the amount of light that hits each sensor is stored. Each individual point on the sensor only measures the amount of photons that have hit it (effectively luminosity). Filters above the sensor result in color information that the "pixel" is capturing.

Image processing is applied to the raw data to determine full color pixels instead of a mesh of red, green and blue pixels. The color space is also reduced to what can be displayed on most screens (which are generally limited to 8 bit color or less).

There is no particular advantage to a sensor shooting in raw that isn't also present when shooting JPEG since the only difference between the modes is that in JPEG shooting, the RAW data is processed in camera, while when shooting RAW, the raw data is stored for later processing.

There are many ways that one sensor can be better than another. The technology of the sensor can differ, CMOS vs CCD are two different technologies for image sensors. CMOS tend to be cheaper and higher resolution, but they sample by scanning. CCDs on the other hand sample frame at a time, but are more expensive and generally lower resolution, but sampling the entire frame at the same time makes them better for video for example.

Additionally, there is the actual resolution of the sensor which should be self explanatory. There is the level of noise in the sensor. The less noise a sensor experiences, the cleaner the signal will be. There is also higher levels of sensitivity which can allow for use in lower light environments and for finer detail to be resolved.

The pattern of the red, green and blue filters also has an impact on the color accuracy and gamut that the camera can support as well as to the level of accuracy in the color. Some sensors layer sensors with different filters which reduces the amount of light that they can work with, but also produces slightly higher resolution from resolving each color from a single point instead of multiple nearby points.

The size of the sensor itself and the "pixels" on it impacts things like diffraction limiting which limit the maximum resolution of the camera due to physical properties of light itself. It also impacts the depth of field and apparent focal length for any given lens. (An APS-C sensor takes up less room, so it crops the image circle thrown by a lens compared to what the lens would throw for a full frame sensor.)

I'm sure there are other ways that sensors can differ in quality that I'm not thinking of, but it should give you an idea of what some of the differences are. Hopefully it also helped clear up the confusion about what RAW photography actually is.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.