Our Dynamic Range measurement system involves shooting a calibrated Stouffer Step Wedge (13 stops total range) which is backlit using a daylight balanced lamp (98 CRI). A single shot of this produces a gray scale wedge from (the cameras) black to clipped white (example below). Each step of the scale is equivalent to 1/3 EV (a third of a stop), we select one step as 'middle gray' and measure outwards to define the dynamic range. Hence there are 'two sides' to our results, the amount of shadow range (below middle gray) and the amount of highlight range (above middle gray).
To most people highlight range is the first thing they think about when talking about dynamic range, that is the amount of highlight detail the camera can capture before it clips to white. Shadow range is more complicated; in our test we stop measuring values below middle gray as soon as the luminance value drops below our defined 'black point' (about 2% luminance) or the signal-to-noise ratio drops below a predefined value (where shadow detail would be swamped by noise), whichever comes first.
Picture Style options
The D3S features the now standard Nikon 'Picture Style' presets, but it has fewer presets than models lower down Nikon's range; in fact it only has four, including 'monochrome'. The three color Picture Styles - Standard, Vivid and Neutral - offer different gradations of saturation and contrast, with neutral being lowest, and vivid being highest. All four picture styles can be fine-tuned, and new styles can be loaded via Nikon's Picture Style Editor, which is built into View NX.
It is no surprise to see that the Neutral Picture Style preset gives the highest dynamic range (albeit not by much in practical terms) as a result of its lower contrast tone curve. The higher contrast 'Vivid' style gives the narrowest dynamic range, but the deep black shadows and clean bright highlights give images shot at this setting a 'punchiness' that suits some subjects. The Monochrome Picture Style offers virtually the same dynamic range as the color 'Standard' setting. In all cases the highlights clip at the same point.
Nikon's Active D-lighting system modifies metering and adjusts contrast at a local level in order to maximize the amount of dynamic range information squeezed into the JPEG file. Because the system operates by analyzing the scene to be captured, this studio test may not represent exactly what you'll see in 'real world' images. It does, however, show very clearly how ADL pushes and pulls the JPEG tone curve to extract the maximum tonal information from shadows and highlights at its different settings.
Notice how with ADL turned on 'Extra High' the tone curve is visibly further to the left, indicating increased brightness in the shadow and midtone areas. In 'real world' images the effect varies from scene to scene, but outside of our rather artificial studio setup we would expect a noticeable retardation of highlight brightness.
ISO Sensitivity and Dynamic Range
The base sensitivity of the sensor used in the D3S is ISO 200. There is also a Lo 1.0 mode that attempts to mimic ISO 100 but it's effectively just ISO 200 over-exposed by a stop. The result is that the sensor becomes saturated and clips to white quite easily, limiting that mode's dynamic range. For most applications, you'd be better off buying a neutral density filter if you need slower shutter speeds than ISO 200 will allow. The same goes for the older D3, and the D300S.
Beyond ISO 200 the highlight dynamic range remains a fairly stable 3.6 (approx) stops above middle gray across the span of the most-used ISO sensitivity settings. This is very similar to what we'd expect from the D3 and the D300s. The shadow range appears to increase then decrease higher up the ISO scale but this is most likely to be a result of noise and noise reduction changing our cut-off point.
|Sensitivity||Shadow range||Highlight range||Usable range|
|Lo 1 (ISO 100 equiv).||-4.9 EV||2.6 EV||7.6 EV|
|ISO 200||-4.7 EV||3.6 EV||8.3 EV|
|ISO 400||-4.7 EV||3.6 EV||8.3 EV|
|ISO 800||-4.7 EV||3.6 EV||8.3 EV|
|ISO 1600||-4.8 EV||3.6 EV||8.3 EV|
|ISO 3200||-4.8 EV||3.6 EV||8.5 EV|
|ISO 6400||-4.7 EV||3.7 EV||8.3 EV|
|ISO 12800||-5.2 EV||3.6 EV||8.8 EV|
|ISO 25600||-5.4 EV||3.7 EV||9.0 EV|
|ISO 51200||-3.4 EV||3.7 EV||7.1 EV|
|ISO 102400||-3.0 EV||3.6 EV||6.6 EV|
Dynamic Range compared
The D3S, along with the D300S, Canon EOS 1D Mark IV and most other high-end DSLRs, offers a JPEG dynamic range that represents benchmark performance in the market at this point in time. Around 3.5-3.8EV of highlight dynamic range from default JPEG output is pretty as good as its gets at the moment from DSLRs (the Sony A900 gives the current best measured highlight range at 4.2EV but the difference isn't enormous in any practical sense), and a total dynamic range of between 8.3-8.7EV means that JPEG from all four cameras respond well to post-capture tonal adjustment when coupled - naturally - with accurate metering.
As you can see from the graph, the 1D Mark IV offers fractionally wider JPEG dynamic range than the D3S overall, but shows a steeper curve at the highlight end of its scale. It is worth remembering though that how much latitude you actually get in highlight areas is dependant in no small part on where the camera (or the photographer) chooses to position the midtone.
|Camera (base ISO)||
|Nikon D3S||-4.7 EV||3.6 EV||8.3 EV|
|Canon EOS 1D Mark IV||-5.1 EV||3.5 EV||8.7 EV|
|Nikon D3||-4.7 EV||3.9 EV||8.6 EV|
|Nikon D300S||-4.6 EV||3.8 EV||8.4 EV|
|Canon EOS 7D||-5.0 EV||3.3 EV||8.3 EV|
|Sony Alpha A900||-5.1 EV||4.2 EV||9.4 EV|
The wedges below are created by our measurement system from the values read from the step wedge, the red lines indicate approximate shadow and highlight range (the dotted line indicating middle gray).
- 18 Photographic tests (Noise)
- 19 Photographic tests (DR)
- 20 Photographic tests (DR)
- 21 Photographic tests
- 22 Movie Mode
- 23 Compared to
- 24 Compared to (JPEG)
- 25 Compared to (JPEG)
- 26 Compared to (JPEG)
- 27 Compared to (RAW)
- 28 Compared to (RAW)
- 29 Compared to (RAW)
- 30 Compared to (Higher ISO)
- 31 Compared to (Resolution)
- 32 Compared to (Resolution)
- 33 Conclusion
- 34 Samples