Nikon D700

Where the Nikon D700 shines is making the most of every available photon in areas dark enough to make lesser camera shutters shudder, minimizing image noise in the process. Some of the other test results for this camera come with caveats, though. Shutter lag, for example, is infinitesimal if the camera’s already focused, but a full-travel shutter press without prefocusing on our full-computer-screen digital stopwatch produced unexciting results. We also discovered that, in certain tests, the faults we found were caused by defaults: specifically, the default Picture Control setting of Standard. As explained below, this mode produces handsome pictures, but sacrifices color accuracy in the process.

Color (9.32)
In our lab testing, our primary concern is accuracy: making sure that the shades reproduced by the camera match what we actually saw when we took the shot. We’d been shooting with the D700 in the field for some time before we started shooting test charts, and eyeballing our test images produces consistently appealing colors, albeit a tad warm. When we started shooting the standard Gretag Macbeth test chart under studio lighting and running the resulting images throught Imatest image analysis software, though, the camera received mediocre scores. After some discussion, we realized that this was the same effect found when testing high-end Canon cameras: the default image settings produce so-so color accuracy, but switching to Faithful mode for the Canons results in a dramatic improvement. Same thing here: switching from Standard to Neutral mode when shooting with the Nikon D700 moved the camera’s color accuracy score from a mediocre 6.89 to a far more palatable 9.32.

In the Imatest chart below, the color captured by the camera is shown in the outer rectangle, the luminance-corrected color is shown in the middle square and the original color chart hue is shown in the small inset rectangle.

 
The ideal color segment indicates the original color chart value,
the outer rectangle is the D700 result.

A second Imatest chart displays graphically just how close the captured values come to the original colors. In this case, the ideal color is represented by a square, the color as captured by the camera is shown in a circle, and the line between them indicates the distance between the two values: shorter lines equal better color accuracy.

Nice short lines in this chart indicate the D700 reproduced
color values accurately.

Resolution (10.27)
Assuming that a high number of megapixels in the camera spec equals the superior resolution performance in your actual photos is a rookie mistake. Optics, sensor quality, and image processing, among other factors, have a powerful influence on photo sharpness. In our testing, we shoot an industry-standard resolution chart at a variety of distances and camera settings, under bright studio lighting, to find the top resolution performance the camera is capable of delivering. Imatest software analyses the test shots and grades them based on the number of alternating horizontal and vertical lines the camera can clearly reproduce (represented as line widths per picture height, or lw/ph) before the chart figures turn into grey mush.

A full-resolution section of a D700 image from an industry-standard resolution chart

The D700 did well in resolution testing, topping out at 1787 lw/ph horizontal with 10.3 percent undersharpening and 1771 lw/ph vertical, with 12.4 percent undersharpening. This relatively high level of undersharpening found in Imatest’s analysis actually works in the photographer’s favor, since images have enough inherent data to respond well to sharpening tools (such as unsharp mask) during image editing.

A resolution score of 10.27 puts the D700 virtually on a par with Canon’s EOS-1D Mark III, the kind of result we’d expect. Only the surprisingly strong showing by the Nikon D300 exceeds the D700’s performance.

The D700 did well in the dynamic range test, especially if you consider the extraordinary breadth of the ISOs being tested. Ordinarily dynamic range falls off as ISO increases, which means a camera that goes beyond typical 1600 or 3200 ISO values, and doesn't offer an official ISO 100 setting either, is at some disadvantage statistically. At the most typical ISO settings, though, the D700 turned in a fine performance. There is an anomaly in our testing results, though, as shown in the graph above. We expect noise reduction to cut down on dynamic range, since data is being removed from the image in an attempt to cut down on speckles and imperfections, a process that ordinarily blurs dynamic range steps as well. In this case, we found apparent improvement in dynamic range from about ISO 2000 on, and this improvement was consistent throughout the standard ISO range (the values actually came back to be nearly equal when we pushed beyond the official ISO range into Hi0.3, Hi0.7, Hi1 and Hi2. We have no scientific explanation for this unexpected results, and have contacted Nikon to ferret out what caused it. We will update this review when we have an appropriate answer.

As for dynamic range scoring, the Nikon ranked slightly below the other cameras here but, again, the lack of ISOs below 200 and the inclusion of those beyond 1600 should be factored in.

Nikon D700 Dynamic Range Scores

White Balance (6.29)
Sometimes looks are deceiving. Whether you look at a white sheet of paper in sunlight, or under a household lamp, or under fluorescent lighting, it looks like a white sheet of paper. That’s because the powerful meat computer between your ears skillfully adjusts for the very different colors of light these sources of illumination actually cast. Our digital cameras have to make a similar calculation to adjust for the color of light when taking photos, with greater or lesser success. And when white balance is significantly off, it’s not a subtle flaw you can easily overlook: just think of all those orange, overly warm shots you’ve seen over the years, taken indoors under incandescent bulbs with a camera that just couldn’t make light of the situation.

We test white balance under a variety of lighting conditions: flash, fluorescent, shaded daylight and tungsten. In each case we shoot the Gretag Macbeth color chart two ways: one using the camera’s automatic white balance system, the other by choosing the appropriate white balance preset. The D700 offers a variety of fluorescent presets, based on the many different types of fluorescent bulbs on the market. We shot with all the presets and went with the best results. Its also worth noting that, for each preset, the camera allows users to finely adjust the settings by moving a cursor across an on-screen color chart. If you take a shot and don’t like the results, it’s fast and easy to tweak the white balance settings and keep shooting until you get the look you’re after.

Our standardized white balance testing procedure calls for shooting in default mode, which in the case of the D700 led to lower-than-expected scores. The default color setting for the camera is Standard, which enhances colors subtly to provide a more visually appealing image. However, as shown in our Color test above, the Neutral setting favors accuracy, producing higher scores when Imatest analyzes the image. To achieve more realistic white balance results, setting the camera to Neutral is recommended.

Auto White Balance (6.91)
The automatic white balance system did a good job while shooting with flash and under fluorescent lighting, but coped badly when asked to accurately capture colors under shaded daylight or tungsten lighting conditions.

Preset (5.67)
Here again, flash performance was excellent, and using the preset resulted in a major improvement when shooting under tungsten lighting. However, none of the multiple available presets outshone the results we found using the auto white balance setting when it came to fluorescent lighting.

Exaggerated White Balance Errors (Presets)

Flash Preset WB - Flash Illumination

Florescent Preset WB - Fluorescent Illumination

Cloudy Daylight preset WB - Daylight illumination

Tungsten preset WB - Tungsten illumination

It's interesting that the Nikon D700 and the Canon EOS-1D Mark III show similar, relatively unimpressive scores compared to their less expensive cohorts. In both cases, the standard color modes aren't the ones that emphasize color accuracy, which throws off their scores on tests that stress precision rather than aesthetics.