This question already has an answer here:

So, I was searching for the exact mechanism behind the workings of ISO (or film speed) in digital camera. I found this video and it was pretty simple. But i was wondering if this video is accurate and if anyone could add extra details to it or correct any mistakes (if any).

P.S. The video didn't have comments enabled either so, i couldn't really tell.

marked as duplicate by Philip Kendall, Saaru Lindestøkke, scottbb, Hueco, mattdm May 17 at 20:07

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.


The video explanation of ISO is pretty good but not perfect. It concentrates on an individual pixel (picture element). The video portrays red, green, and blue light, playing on a discrete pixel. In reality, each pixel on a modern digital color camera is subdivide into three sub-pixels, one for red, one for green, and one for blue. These are the three light primary colors. We can image and display color pictures if we record the intensities of these primary colors.

Each of the three sub-pixels sites are covered by a strong transparent filter. The filters are red, green, and blue colored material. Thus each sub-pixel receives 1/3 of the incoming light. Let me add, the distribution of red vs. green, vs. blue sensor site is not equal. The distribution is uneven to force the image sensor to mimic the human eye as to sensitivity to color. Another point, some futuristic sensors have a scattering of unfiltered sites. This method boots the overalls sensitivity to light. Software thus concludes the color of the light playing on an unfiltered sites.

When the shutter opens, each site is bombarded by photon hits. These are in direct proportion to the colors and intensities of the vista being photographed. The photon hits generate an electric charge inside the photo site. The more hits, the greater the charge. If the vista is poorly lighted this will be feeble light, charges on the sensor sites will be super tiny. If the vista is strongly lighted, the charges will be greater but still too tiny to be of much value. Thus each site contains an amplifier. The job of the amplifier is to boost the strength of the charge to make it useful. Each amplifier has its own characteristics thus each will boost the charge differently. This is the major cause of what we call noise. This is akin to static on an audio system. The higher we turn up the volume, the more the static / noise.

ISO (International Standards Origination based in Geneva) is the recognized authority on the sensitivity of photographic film. They develop the test methods uses to assign a sensitivity value ISO number. Digital imaging started as a sub-set of chemical imaging (film). The digital community roughly follows the ISO guidelines when it comes to assigning sensitivity values (ISO).

Noise in digital imaging creates artifacts (nonexistent image content). There are many types. Noise is a granularity of the image akin to a similar artifact of film called “grain”. When working in feeble light, we turn up the amplification to allow imaging under this adverse situation. As we turn up the amplification, the ratio of static / noise increases thus the image displays the artifact we call noise.

  • What you are calling "sub-pixels" are simply photosites (a/k/a pixel wells). There are not three subpixels on the sensor for each pixel in the resulting image file. A typical 20 MP sensor has 10 million photosites filtered with "green", 5 million photosites filtered with "blue", and 5 million photosites filtered with "red" (which is usually somewhere between greenish yellow and orangish yellow, not "red"). Demosaicing interpolates a red, green, and blue value for each photosite that results in 20 million pixels with Red, Green and Blue channels. Display technologies do use three subpixels. – Michael C May 20 at 0:27
  • So from what i understand, The video explains the mechanism while increasing the brightness of an image when it (the image) is edited after the photo was taken (in photoshop or lightroom or any software like that). Whereas, what you have explained represents the mechanism during the change in ISO setting of the camera. – Zinc May 28 at 13:56

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