No announcement yet.

Measuring Dynamic Range

  • Filter
  • Time
  • Show
Clear All
new posts

  • Measuring Dynamic Range

    I've been wondering about this for quite some time. With all the talk about dynamic range and its importance, I was wondering if it could be quantified within the software somehow; in other words, how much more range did I get with this RAW file than from a jpeg file of the same image? Can dynamic range even be measured? I've heard that you get more dynamic range by taking RAW images but I'm hardly seeing much difference between RAW and jpeg in my photos. I know the RAW file size is much larger before processing, but that doesn't really show me dynamic range difference, so I'm wondering if there's anything in ACDSee that would identify DR for a given image. Thanks for any input!

  • #2

    In a digital image, dynamic range can be simply described as the ratio between the brightest and darkest parts of an image, from pure black to brightest white, and in ACDSee can be graphically seen in the histograms.


    With regard to RAW versus JPG, there is a plethora of information on line. For example

    You should perhaps also read up on HDR (High Dynamic Range) images. HDR is a technique used in digital image processing to combine a series of images with different exposures to produce one image with an optimal overall range,
    Last edited by Greyfox; 01-10-2023, 06:02 PM.


    • #3
      Many thanks for your explanations and links Greyfox, but I've never seen a histogram with numbers, so how can I develop a ratio between the lightest and darkest areas?
      Last edited by jimlock99; 01-11-2023, 07:01 AM.


      • #4
        Originally posted by jimlock99 View Post
        ..I've never seen a histogram with numbers, so how can I develop a ratio between the lightest and darkest areas?
        I'm not sure why you want to have a actual number to indicate the dynamic range of any particular image, but with the histogram for an RGB image, you can consider the left hand end as 0, and the right hand end as 255. The histogram shows the spread of the pixels in the image across that range. If it is hard up against either end, it may in fact be clipping, in which case some information is being lost.



        • #5
          Thanks Greyfox, those links had some interesting and helpful info. Since dynamic range is discussed and referred to so much, it would be nice to actually quantify it through a measurement process of some sort. In other words, when does an image have a high dynamic range as opposed to a "normal" or regular dynamic range? I guess if you were to compare two histograms of the same scene, one with normal exposure and the other with HDR exposure applied, you could easily spot the HDR histogram (higher mountains?), but at what point does the image change from normal exposure to HDR exposure?


          • #6
            Originally posted by jimlock99 View Post
   what point does the image change from normal exposure to HDR exposure?
            I don't know that there is a definitive answer to that.

            The optimum dynamic range of an image would be where the image includes all of the tonal range of the scene that was photographed, and that would be different for different scenes. (What is the dynamic range of a photo of a black wall, or a white wall?)

            There is also different ways of taking photos when dynamic range is a prime consideration.
            There is what some refer to as "One shot" or "Single Exposure" HDR, and there is "Bracketed shots HDR".

            That article describes single shot HDR, its advantages and its disadvantages, and how to go about it using "ETTR" (exposing to the right)
            With single shot HDR particularly the dynamic range of the camera body itself is important.

            For bracketed shots HDR, See
            That article describes the basic method of taking the shots. Note that In some cameras there is provision to have the merging of the shot done in camera, otherwise it can be done in post processing.

            I'm not sure there is a simple way to quantify that, or if there is what practical use it would be, but then I'm not a professional photographer.

            For what it is worth, Affinity Photo does provide some statistics with its histogram, Mean, Std Dev, Median, Pixels.

            Click image for larger version  Name:	Hystogram with values.jpg Views:	0 Size:	34.2 KB ID:	63592
            The greyed out items on the right become active and give the values at the position under the cursor when it is placed on the histogram.
            Last edited by Greyfox; 01-12-2023, 05:07 PM.


            • #7
              Originally posted by jimlock99 View Post
              at what point does the image change from normal exposure to HDR exposure?
              It's unquantifiable, because there's no formal definition of what "High Dynamic Range" means. It's as much a marketing term as anything, and is meaningless in objective, measurable terms - which is precisely why its so popular with fanboy owners of certain brands of camera, who have incorrectly co-opted the term "dynamic range" to refer to boosting shadows at low ISO without too much of a noise penalty.

              As Wikipedia says, it's:

              a dynamic range higher than usual​​
              But then, what's "usual"?

              It's a subjective concept, Jim. Hell, Capture One even uses the term to describe its Shadows/Highlights slider panel:

              Click image for larger version  Name:	C1_HDR.jpg Views:	3 Size:	9.1 KB ID:	63706

              Does that mean that every time I use any of these sliders on an image (and I use them all the time), I've created an "HDR" image?

              "Yes" and "no" are equally legitimate answers - because it's an essentially meaningless phrase.

              This is a useful reference:

              Recently I’ve been pondering which camera to shoot an upcoming project on, so I consulted the ASC’s comparison chart. Amongst the many specs compared is dynamic range, and I noticed that the ARRI Alexa’s was given as 14+ stops, while the Blackmagic URSA’s is 15. Having used both cameras a fair bit, I can tell […]

              (It isn't primarily about stills camera dynamic range, but the points made are 100% transferable.)

              As you'll see, it's clear that there's significant subjectivity and flexibility of terminology around dynamic range; and it concludes that:

              Whereas [another metric] at least has a standardised test, there’s no such thing for dynamic range.
              I'll close with a bit of silliness - but it makes a point nevertheless.

              This is a typical stepped zone chart:

              Click image for larger version  Name:	Zone_Scale-585x80.jpg Views:	0 Size:	8.8 KB ID:	63710

              It contains everything, from 0,0,0 to 255,255,255.

              If you took a photo of it with your favourite camera, you'd expect to be able to see it all, right?

              How much more "dynamic range" can there be? That's everything right there!

              Attached Files
              Last edited by KeithReeder; 01-21-2023, 12:37 PM.


              • #8
                Thanks to both Greyfox and KeithReeder for your very informative, thoughtful, and concise posts about this topic. I've read all the links you both provided and come to the conclusion there just "ain't no such thing" as HDR, no matter who's using the term. I prefer to believe that some images will have a greater dynamic range than others, which may or may not make the image more appealing, and it is something worth striving for with certain images. To quantify that with any kind of measurement would be foolish and counter-productive.


                • #9
                  I prefer to believe that some images will have a greater dynamic range than others
                  Yep, that's certainly true, Jim.

                  For any image, there will a darkest dark, and a lightest light: the distance between the two is that image's DR. Simply making the shadows/blacks a bit darker and/or the highlights/whites a bit brighter, will extend the end-points of the DR​ of that image.

                  As you say, whether that makes for a better image is a matter of debate: personally I'm not a fan of blocked shadows in my bird and wildlife photography, so I'll usually lift them enough to (for example) reveal plumage detail on the underside of the bird; and I really don't like overly-bright highlights, so I'll routinely knock them back somewhat too.

                  So I'm all over the DR of the image as related to the RAW data out of the camera, but I stop well short of what people refer to as "HDR" - that garish, hyper-real (and not at all realistic), cartoonish look that was so popular a while back (this is a free, publicly available wallpaper, so it's OK to post it here):

                  Click image for larger version  Name:	HDR.jpg Views:	0 Size:	193.6 KB ID:	63758

                  But it's a fine line: I like an image to "pop", rather than just be a flat representation of a bird or beasty, and to some people, this would qualify as "HDR":

                  Click image for larger version  Name:	C17E_IMG_7334.jpg Views:	0 Size:	497.3 KB ID:	63759

                  It isn't - it's just good light, and my processing style (and yes, it does cover a pretty wide DR) - but that's the ongoing point: there's no one, definitive meaning of the phrase under discussion.
                  Last edited by KeithReeder; 01-25-2023, 08:54 AM.