I have a trial copy of RawDigger installed, and I understand that for my purposes I should shoot the same subject at different apertures to see if the SNR decreases at faster settings. But how exactly do I read the screen in RawDigger to evaluate the results? I see various acronyms and numeric values in the attached screenshot, but they are meaningless to me. (Note, the photo is just an example; I know it's not the right kind of image to use for this.) Sample RawDigger display

  • 2
    \$\begingroup\$ Possible duplicate of Measuring SNR of a DSLR from raw image \$\endgroup\$ Commented Jan 6, 2019 at 17:48
  • \$\begingroup\$ @RomeoNinov I don't see how this is a duplicate. The other question asks "how do I measure SNR?", and is answered with "use software, for example RawDigger." But how to use RawDigger? There might be an argument that questions asking aobut the basic use of software, such as RawDigger, are off-topic for Photo.SE. But I don't see how this is a dupe. \$\endgroup\$
    – scottbb
    Commented Jan 13, 2019 at 17:37

1 Answer 1


Before you can understand what RawDigger is showing, you need to understand a few basics.

  1. Most camera sensors are actually colorblind — they only detect light intensity, not the color of light. In order to record color information, individual pixels are filtered to be more sensitive to hues in the red, green, or blue wavelength regions, thus making each sensor pixel represent either mostly "red", "green", "blue".

    The individual color-filtered pixels are arranged such that any 2×2 sample of sensor pixel data will contain 1 red pixel, 1 blue pixel, and 2 green pixels. That is, there are twice as many mostly-green-sensing pixels on a sensor than there are mostly-red and mostly-blue pixels. This is arrangement is known as a Bayer filter (Wikipedia) (there are other schemes in existence, but the majority of digital cameras follow this scheme). To generate output images (such as JPEGs), the RAW sensor data needs to be interpreted according to how the red, green, and blue pixels are laid out for that particular camera model. This is known as demosaicing (Wikipedia).

    Related questions:

  2. The signal-to-noise ratio (SNR) of a digital image (or region of an image) is defined as the average value of the pixels in the image (region) divided by the standard deviation of the pixel values. Standard deviation is a measure of how spread out the individual samples in a dataset are, compared to the mean, and is denoted by 𝜎 (the greek letter "sigma"). It is calculated as the square root of the mean distance from the sample set's average value — the root-mean-square deviation.

So now, we can understand what RawDigger is showing.

  1. The first section of the screenshot is showing some metadata from the image's EXIF information (camera model, lens, focal length, exposure settings, etc.).

  2. The next section, titled "Image 6016x4016", is showing per-color-channel (red, green, blue, and 2nd green (as described above)) information over the entire image:

    • minimum and maximum values recorded. In your example, [R,G,B,G2] of [0,0,0,0] means your camera recorded "pure black" — that is, as dark as your camera can sense.
    • average value. In your example, the average red pixels recorded less light than the blue pixels, and the green pixels recorded more than both red and blue.

    • the root-mean-square deviation. In your example, a typical red pixel varied from the mean (of 1176) by 2065. The SNR is therefore 1176 / 2065 = 0.57. The green, blue, and green2 SNRs are similar (0.52 – 0.54).

    Incidentally, the title of the section is describing the width (6016) and height (4016) of your image.

  3. The next section, had you selected a region in the image, would show the [R,G,B,G2] min, max, average, and RMS deviation for that region, rather than the entire image.

  4. The next section, titled "5529:4029", is showing the [R,G,B,G2] values for the pixels under your cursor.

  5. The next section sets display mode options.

  6. The last section, "OvExp/UnExp Stats, shows the per-color channel over- and under-exposed statistics for the image. In your case, it shows 7–12% (depending on the color channel) of the pixels in your image are underexposed.

The screen information can be found in the RawDigger user manual.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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