Nikon D300 Digital Camera Review

We found that the Nikon D300 performed extremely well overall in our tests; the color in captured images was good, and there was a great level of detail. But it wasn’t without its quirks; the long exposure noise reduction made only a very slight difference and takes a long time. But the iamge quality is overall extremely good.

One important caveat to consider here: with SLRs, we would normally test with the kit lens that is sold with the camera, but the D300 does not have this option. Instead, we tested it with a Nikkor 17-55mm AF-S lens that added $1200 to the cost. The total cost of the lens and body combination that we tested was about $3000.

Color (9.04)
The images that the D300 captured had bright, accurate color. Images came out looking as you would expect from a $3000 camera, with color that was very close to the real world.

We test color accuracy by photographing a color test chart under precisely controlled lighting conditions, then running the images through Imatest, a high end image analysis program. This looks at how close the colors in the image are to the originals on the chart, and produces this version of the test chart.

In this, the outer section of each of the color is what the camera captured, while the inner square is the camera captured color after it has been processed with luminance correction. The vertical rectangle represents the original color.

Imatest also produces a graph that illustrates the accuracy of the colors the camera captures; the squares are the original, ideal colors, and the circles are the colors in the captured image. The shorter the line between the two, the more accurate the camera is at capturing colors.

Most of the lines here are short; the colors are generally pretty accurate. The exceptions are some of the blues and some of the more subtle reds; these colors are a little off, but these are minor issues. The overall mean color error was an impressive 6.63, and the images were only about 6% oversaturated. Both of these are very good figures, showing that the D300 captures good, accurate color. These results show a slight improvement over the Nikon D200, but the color performance is not significantly changed. But they aren’t as good as some similar cameras that we’ve tested: Canon 1D Mark III performed slightly better in this test, managing a mean color error of just 5.69. Likewise, the Canon EOS 40D also had very slightly better color; we found a mean color error in our tests of 6.71.

We tested the D300 in its default picture control mode (called standard), but it also offers neutral, vivid and monochrome picture control modes. We also tested in the default color space setting (sRGB), but the D300 is unusual in also offering support for the Adobe RGB color space model. This supports a wider range of colors, which would make the images more suitable for post-processing. However, most users won’t need to use this setting; only those who heavily edit their images and want the extra color space it offers will need this.


Resolution (11.89)
With a 12.3 effective megapixel image sensor, it’s no big surprise that the D300 did well in our test that analyzes the resolution that the camera can capture. But there’s more to this test than the image sensor; we’ve seen plenty of cameras with high megapixel counts, but poor resolution because the image wasn’t processed well after it was captured. To analyze this, we take a series of photos of a resolution chart at a variety of settings for focal length and aperture, then run the captured image through Imatest to see how much detail there is in it. Our scoring system takes resolution into account, so a 5-megpixel camera can earn the same score as a 10-megapixel one if the images it captures contain the same level of detail relative to the resolution.


Imatest produces a measure from this analysis called line widths per picture height, which is a measure of how many alternating black and white lines the camera could capture before they disappear into a grey mush. The higher this number, the more detail the camera is capable of capturing and storing in an image. The D300 did extremely well in this test; the combination of the D300 and the 17-55mm lens we used for testing managed to capture an impressive 2031 lw/ph horizontally and 1762 lw/ph vertically, both with undersharpening of about 3%, which is pretty close to the 4% ideal we look for in digital SLRs. These are the highest results we have recorded for SLRs, so it’s safe to say that the D300 captures a great level of detail, and that the images would be able to be enlarged to huge sizes, or cropped in tightly and still retain plenty of detail.



Noise – Manual ISO (13.25)
Noise is the stuff that ruins photographs: the blotchy patches in images and the static that turns a family snapshot into a poorly tuned TV picture. And more megapixels often means more noise; as the size of the individual sensor elements gets smaller, they become more prone to picking up the electronic background noise that causes noise in photos. And upping the sensitivity ot light of the camera by increasing the ISO just adds another problem; as the ISO goes up, the noise gets amplified along with the signal from the light hitting the sensor. So, given that the D300 takes photos at an impressive 12 megapixels at an ISO setting of up to 3200, how much noise is there in the photos it takes? The answer is surprisingly little; we found that the images were impressively low in noise.

We test noise by taking a photo of our color test chart under bright light, then gradually increasing the ISO level and using Imatest to analyze the noise in the resulting images. We then produce a graph like the one below that shows how the noise increases as the ISO increases; the noise is measured as a percentage of the total signal, and the higher the noise, the worse the image.

The D300 has three different levels of noise reduction; low, normal and high. There’s also a setting to turn off the noise reduction completely, which is indicated by the pink line. As you can see, the noise reduction features did an excellent job of reducing the noise in images; even at the maximum setting of ISO 3200, the noise was only just above 1 per cent with the normal and high noise settings. The default normal setting also did a good job of keeping noise down; although the noise at 3200 ISO was noticeable, it was a lot less obvious in the images that many cameras we’ve tested.

There is still noise in the images with the High ISO Noise Reduction setting on high but it is much reduced from the normal setting. However, images did lack some definition with the high setting enabled; although the camera is doing a great job of preserving the color information in the image, some of the detail gets slightly reduced by the noise reduction process, giving images a slightly soft look; see the edges of the text in the images above for an example. We’ll go into more detail on this in the low light tests below.


Auto Noise (8.35)
We also test the noise levels with the ISO set to automatic, where the camera picks the appropriate ISO level for the lighting situation. Because this is a well-lit color chart, the camera unsurprisingly picked the ISO 200 setting, which produced a low level of noise.



White Balance (9.17)
In this test, we look at how well the camera judges the qualities of the light it is capturing; particularly the color temperature. White light often isn’t as white as you think; the color of the light emitted by the sun is very different to the color of the light given off by a florescent light. This is called the color temperature. Your eye and brain automatically adjusts for the color temperature of the light, making white object seem white whatever the light. The camera has to as well, with something called white balance. We test two aspects of this; the automatic white balance (where the camera examines the light and adjusts the white balance) and the presets, where the camera has a number of built-in settings for different types of lighting. We test by illuminating a color chart in 4 different types of light (flash, florescent, shaded daylight and tungsten), taking photos using both the auto and preset white balance settings, then analyze the captured images in Imatest.

Auto (6.7)
As with most SLRs, the D300 had some issues when using auto white balance: it had issues with several of our light sources; particularly the flash and tungsten settings. The latter of these was particularly off; the whites had an unpleasant orange tone. The daylight white balance was acceptable, though; it did a decent job of judging the light in that situation.

Exaggerated White Balance Errors

Auto WB - Flash Illumination

Auto WB - Florescent Illumination

Auto WB - Daylight illumination

Auto WB - Tungsten illumination

Preset (11.6)
The white balance presets were much more accurate; in this test, the images had white balance that was accurate for most of the light sources; the tungsten and overcast daylight preset were almost perfect.

Exaggerated White Balance Errors (Presets)

Flash Preset WB - Flash Illumination

Florescent Preset WB - Florescent Illumination

Cloudy Daylight preset WB - Daylight illumination

Tungsten preset WB - Tungsten illumination

One thing that we don’t score on is the evaluative white balance, where you supply the camera with a white object and it judges the color temperature of the light from that. We found that this mode on the D300 did an excellent job of judging the light; it produced much more accurate results than the auto white balance. So, the bottom line here is that if accurate white balance is important, use either the presets or the evaluative white balance. The auto white balance did an adequate job of judging daylight, so it should be fine for outdoor shooting, but it didn’t work well under tungsten of florescent light.

Still life

Click to view the high-resolution images.

Still Life Scene
ISO 200	ISO 200
ISO 250	ISO 250
ISO 320	ISO 320
ISO 400	ISO 400
ISO 500	ISO 500
ISO 640	ISO 640
ISO 800	ISO 800
ISO 1000	ISO 1000
ISO 1250	ISO 1250
ISO 1600	ISO 1600
ISO 2000	ISO 2000
ISO 2500	ISO 2500
ISO 3200	ISO 3200

Low Light (9.02)
In many situations, you can’t use the flash that comes built into your camera. Flashes are banned at most performances, so if you want to capture your childs oscar-worthy performance as a sunflower in the school play, you’ll have to use whatever light is available. That’s why we test the low light performance of cameras; to see how well they can deal with this sort of difficult lighting situations. We test the low light performance of cameras in two ways, measuring their performance with long exposures and at different light levels. In both tests, we found that the D300 did an excellent job of capturing images without producing ugly, grainy images; they were consistently low in noise, but with strong color.

Our first low light test measures the performance of the camera at different light levels, from 60 lux (about the same as a well-lit indoor room) to 5 lux (about what you’d get in a dark room with just a single candle). For this test, we set the ISO level to 1600. We then analyze the captured images, looking for how accurate the colors remain and how much noise there is in the image as the camera struggles to make the most of the existing light.

Low Light Tests
60 Lux	30 Lux
15 Lux	5 Lux

The D300 did well in this test; as the light level fell, the amount of noise in images grew, but not as much as many other cameras that we’ve seen; even with just a minimal 5 lux of light, the D300 produced images with low noise. With the noise reduction disabled, we measured the average noise at about 2.4 percent, which is a little higher than the noise we saw in the canon 40D (around 1.5 percent). But with the noise reduction on its highest setting, the noise dropped to just 1.11 percent, while the 40D’s noise reduction mode only pushed the noise down to 1.22 percent. So the noise reduction feature on the D300 did a very impressive job here; although the noise in images was very low to begin with, it reduce it even further.

For our long exposure test, we take a series of images with exposure times of 1 seconds to 30 seconds and then use Imatest to analyze the amount of noise present in the images. For the D300, we tested it with the long exposure noise reduction both disabled and enabled. We then produce the following graph, which shows how the noise in the images increases with the exposure time.

What this shows is that the D300 did a great job with keeping the noise level low, even with the longest 30 second exposure. Even with the long exposure light level setting disabled, the noise level never rose above just over 1.2 percent, which is a very small amount of noise. The long exposure noise reduction setting did a very creditable job of reducing this low noise level even further; the noise in this test hovered at around 1 percent. Both of these figures are extremely good; in many compact cameras we see noise levels of 4 to 5 percent, and the D300 did better in this test than most other SLR cameras as well. Of the camera that we’ve tested in the past year, only the Canon 1D Mark III had lower noise in long exposures.

However, there is a price to pay for using the long exposure noise reduction mode; it doubles the shooting time, as the camera takes as long to process the image and reduce the noise as the shutter speed. So, if you are shooting for 15 seconds of exposure time, you won’t see the image for another 15 seconds after the shutter has closed. Whether this is worth it depends on how important the noise reduction is to you; if you are photographing things that don’t mind holding relatively still (such as the moon or the stars), then it’s probably worth it.


Dynamic Range (11.27)
Dynamic range is how well the camera can capture the full range of shades, from the deepest black to the whitest white. While this sounds like a washing powder commercial, it’s a very important test; the wider the dynamic range a camera can capture, the more of the highlights and lowlights of an image they will be able to capture. Again, the D300 did very well here: we found that it had good wide dynamic range, and that this dynamic range was well maintained as the ISO level increased. Although it did drop off, the D300 had less of a steep decline of dynamic range as the ISO level increased.


Most digital SLRs do well in this test, as the extra processing power that they have allows them to keep more shadow and highlight detail in images as they compress the images. The D300 did better than most, though; the only digital SLR we’ve tested that had a wider range was the Canon 1D mark III.  

Speed/Timing – All speed tests were conducted using a 2GB SanDisk Ultra II Compact Flash Card, with the camera shooting large, superfine JPEGs.

Startup to First Shot (9.7)
We test how quickly cameras start up by photographing digital timers. Our first test measures ho quickly the camera can take the first shot after you turn it on, and the D300 was very fast here; we could turn the camera on and snap a shot in around 0.3 seconds. That’s very quick; most SLRs take a couple of seconds to warm up, but the thing that took the longest time with the D300 was turning the dial to turn the camera on. This doesn’t count time for the lens to focus, though; we use manual focus in this test to get the quickest possible time.

Shot-to-Shot (10.2)
Shooting JPEG images on a fast CompactFlash card, the D300 managed an impressively short time of just 0.17 seconds between shots, for an average of 5.8 frames per second. That’s in the continuous high mode; in the slower continuous low speed mode, the time between frames drops to .33 seconds, for an average of 3 frames per second.  In the continuous high mode, we were able to capture 21 frames before the camera began to shoot at a significantly slower frame rate, but it was able to keep shooting in the continuous low mode continuously, until the memory card was filled. This is dependent on the speed of the card, though; we use a fast memory card, so a slower one would be more of a bottleneck. An indicator on the top display shows how many approximate frames of buffer memory are left.

Shutter-Shot (9.0)
We were unable to ascertain the shutter delay of the D300; it was too small to accurately measure. But we can safely say that it isn’t much; the camera is extremely quick to capture an image when you press the shutter.

Processing (9.0)
Image processing was also very, very fast; images were displayed less than a second after they were captured.