What's On The Calendar:
Upcoming Birthdays: alan_n (57), rsimms (34), Clay (65), streetshooter (59), Daniel Diaz (37), kshapero (59), GUER03 (42), fredster (65), James Fitzgerald (49), Hemrik (20), RoNinHeart (51)

Who's Online Now:
262 Guests, 19 Users
joudbren, Joemc, Dallas, stenrasmussen, Marathonrunner, ThomasAdams, dpowell, Fanie, Etienneb, Ulysses, kbr, Turbaaniherra, bguijt, the_traveler, afx, alan_n, rbsandor, simon b, mcananeya

Forum Stats:
Total Members: 3136
Total Posts: 90057
Total Topics: 11791
Total Categories: 5
Total Boards: 41

As most of you already probably know, I’m a practical type of guy. So when it comes to measuring the dynamic range of my cameras, I do it the old fashioned way just like when I measured Velvia and other films in the analogue era (Wow that sounds like a long time ago).

Dynamic range

But what exactly do I mean by Dynamic range? Well concerning my D300, I want to find out the difference in f/stops between the exposure needed to get some noticeable texture in the darkest parts of a scene and the exposure needed to reach pure white or highlight clipping at 255 RGB points. Then I will be in a position to know the maximum luminance range that my camera can record (luminance range being the dynamic range equivalent from the scene to be photographed).

RAW format

I only shoot in the RAW format so I don’t have any clue to what differences this may have on Jpegs or tiffs. I prefer RAW for a few reasons, but one of them is that it provides me with some room to extend the dynamic range and maybe save some highlight detail in an overexposed image or the opposite, save some shadow detail from an underexposed image.

Testing procedure

My testing procedure is simple and I’ll try to be as analytical as possible to help you measure your own system.
Pick a neutral surface as a target, for instance a gray card. I prefer an old beaten up card that has some dimples or other markings on it so it’s not entirely without texture.
Place the card under neutral lighting like daylight that can stay consistent during the whole test (clear sky with no clouds). If you find daylight to be too strong, you can either use a diffuser to lessen its strength or go indoors under tungsten light.
If you choose the second option, you will have to put up with the longer exposures. The different lighting scenarios offer pretty much the same results, so it doesn’t really matter which one you choose. The only difference I have found is that there is a shift of the whole scale towards the highlights, so my +3.6 stops over base exposure become +4 stops. The other advantage with daylight is that you get higher shutter speeds. What is important though is that you set a correct white balance setting. So if possible choose a manual reading off the gray card so RGB values are identical.
Set the camera on Manual exposure mode and choose the base ISO (for the D300 that’s 200 ISO). We set exposure based on a spot meter reading off the grey card. Fix the aperture at f/8 or so – this is not critical since it depends on the strength of the lighting. If you do the test in daylight you may go as high as f/11 or f/16 or as low as f/5.6 indoors.
From this point we usually turn the shutter speed dial 21 steps towards underexposure which will set the camera at 7 stops under (21/3=7). (If you reach the maximum shutter speed, limit your range to this or change the aperture). From here we will start taking the exposures moving the shutter dial a step at a time (1/3 stop) until we reach 21 steps over base exposure. So now we have 43 frames that cover a range of 14 stops. This should cater for most cameras – if your camera needs more frames for the test, just shoot more frames!

Examining the results

Next we examine the results. Pick a light frame and view it at 100% in Nikon view. Toggle between frames till you find the lightest that still holds some detail. Usually there will be a frame with some faint detail across the whole screen and then the very next frame has no detail. This frame with no detail is usually the one I count as my lightest frame. When this frame is opened in Capture NX and we use the exposure tool to dial in minus two stops of exposure correction, then a lot of detail is revealed. You can do the same with later frames and still get detail but we need to set the limit somewhere. Also this frame will measure something in the region of 255 RGB points with some areas dropping to 254 depending on how evenly the card was lit. What is important is that the previous frame gives anything but 254 or 255 values.  

Next pick a dark frame and view it at 100%. Toggle between frames till you find the darkest that still shows some detail. The dark frames are a lot harder to judge since noise will start to distract and fool the eye into thinking it can separate details. Try to focus on a certain texture or marking and as you get into the darker frames watch for when the noise alters the point which you have focused on. I usually pick this frame as my darkest with detail. Some may insist on going darker but that’s up to you. I am concerned with what I can see and use. This frame when opened in Capture NX will reveal a lot of detail especially if the exposure dial is used to dial in plus two stops of exposure correction. You can do the same with later frames and still get detail but we need to set the limit somewhere. Also this frame will measure something in the region of 4 to 12 RGB points. 

The lightest frame that hides some detail and tone will represent one end of the dynamic range. The darkest frame that is just distinguishable will determine the other end. By noting the amount of difference between these exposures, we will know the cameras dynamic range in terms of f/stops. Also the frame that measures closest to 128 will be the middle frame representing base exposure.

Results from my D300

My darkest frame from the 200 ISO test under tungsten lighting was 5.6 stops from base exposure and produced a RGB reading of 4 points. My lightest frame was 3.6 stops from base exposure and produced a RGB reading of 255 points.

That’s a total of 9.3 stops of dynamic range at base ISO. This is the same amount as measured under daylight as well.

For comparison, my D200 had the following results. At a base of 100 ISO the D200 managed 8 stops of dynamic range from -5 to +3. 

Is this really the Dynamic range of the camera?

Well, technically no. But that doesn’t really matter, because in essence we are also measuring the computer system and the human controlling the test. So in effect we are measuring a complete system which happens to be our system! So even if our D300’s were to show the same performance, my results may be worse or better than yours, depending on the quality of my monitor and the discerning ability of my eyes.
 
Happy shooting
John Cazolis