D-Lighting is a shadow & highlight enhancement feature first seen on Nikon's Coolpix range of compact cameras and later on the DSLRs. It is now offered as a shooting setting, with four levels that can be applied (or it can be turned off). Rather than only adjusting the shape of the tone curve (as most similar systems do), the Nikon system also adjusts exposure by one or two thirds of a stop. There is also an Auto mode that selectively applies Active D-Lighting, based on its assement of the contrast in the scene.
D-Lighting's effect can be seen in our dynamic range test, where we've set the exposure (in aperture priority mode) with D-Lighting off, then allowed the camera's metering to adjust the exposure as it would in a real-world situation.
To demonstrate the difference this can make we set up our simple 'shadow scene' and shot it with all the different levels of Active D-Lighting. As can be seen from the histograms, applying any level of D-Lighting lifts the shadows by a standard amount. As the magnitude of D-Lighting is increased, it shortens the exposure and adjusts the top end of the tone curve to correctly represent brightness and retain more of the highlight detail.
|Active D-Lighting: Off||1/13 sec|
|D-Lighting: Low||1/13 sec|
|D-Lighting: Normal||1/15 sec|
|D-Lighting: High||1/20 sec|
|D-Lighting: Extra High||1/20 sec|
In the real world
D-Lighting is supposed to assess the contrast of a scene and attempt to render it in a way that is more consistent with the way that the human eye and brain perceive the world. This means applying different tone curves to different parts of the image, rather than making corrections to the entire image. Here we compare two shots taken with Active D-Lighting turned off and to Extra High.
With D-Lighting turned to its highest setting, it does a better job of retaining highlights by underexposing slightly, then pulling the rest of the image brightness up, while retaining correct color and contrast.
|Active D-Lighting: Off||Active D-Lighting: Extra High|
|F8 1/200 sec||F8 1/250 sec|