Nikon D800 Review

Introduction

As the test scores began rolling in, however, it became clear that Nikon's gambit paid off—for a certain kind of photographer, at least. While the $2,999.95 (body-only) D800 isn't all that fast, its high-resolution sensor provides photographers with flexibility that no other full-frame model on the market can match. It became our camera of choice for capturing many events, with a combination of comfort, control, and image quality that we felt comfortabe relying on.

Design & Usability

{{section_header}}{{section.name}}{{/section_header}}

The D800's 36.3-megapixel sensor is complemented by tons of control and a familiar physical design.

The Nikon D800 offers just about everything the professional or high-end hobbyist photographer could ask for in terms of design. For a full-frame body, it's relatively compact and lightweight (especially when compared to the flagship D4), and it's very very sturdily built. The refined ergonomics, perfected over many generations of Nikon DSLRs, make it comfortable to hold, even for long periods of shooting.

Many of the common adjustments you might make on the fly—white balance, ISO, metering, and drive mode—are all placed on the top plate, on dials to the left of the optical viewfinder. The back of the camera is nearly identical to that of the D700, save for a more compact directional pad and updated controls for triggering live view and video recording.

It didn't take us much time to get used to the D800's handling, and we found that most of the adjustments we wanted to make were immediately available. Some of the changes were more complex than necessary—we're looking at you, custom white balance—but all were easy enough to remember, with dedicated buttons and switches for nearly everything. We're still not fans of the somewhat archaic Nikon menu system, but that's a venial sin when taken in the context of an otherwise sterling design.

{{photo_gallery "Front Photo", "Back Photo", "Sides Photo", "Top Photo", "Bottom Photo", "Lens Photo", "Lens Photo 2", "3D Lens Photo", "Media Photo", "Easy Mode Photo", "Manual Controls Photo 1", "Manual Controls Photo 2", "Zoom Photo", "Zoom Photo 2", "Ease of Use Photo", "Battery Photo", "LCD Photo 1", "LCD Photo 2", "EVF Photo 1", "EVF Photo 2", "Ports Photo 1", "Ports Photo 2", "Ports Photo 3", "Ports Photo 4", "Ports Photo 5", "Ports Photo 6", "Handling Photo 1", "Handling Photo 2", "Handling Photo 3", "Handling Photo 4", "Buttons Photo 1", "Buttons Photo 2", "Buttons Photo 3", "Box Photo"}}

Features

{{section_header}}{{section.name}}{{/section_header}}

The best feature is undoubtedly the high-resolution sensor, though it comes with some drawbacks.

The big headline draw for the Nikon D800 is not a specific mode or option, but simply the hardware that's included for its $3000 price. Look past the camera's durability and its high-resolution 36.3-megapixel image sensor, though, and you'll see a full-featured camera that offers enough for both professionals and enthusiasts. The D800 includes features like built-in timelapse, full HD video, in-camera HDR, in-camera editing, and multi-exposure shooting. These aren't unique to the camera, but they're nice extras that complement the full level of control offered in the D800.

There's still no getting around the fact that 36.3 megapixels is an incredible amount of resolution for a full-frame camera. It puts the D800 closer to most medium format backs than rival full-frame DSLRs, and it's certainly a bold gamble by Nikon. Prior to its release, the common notion seemed to be that sensors had rightfully topped out at around 20 megapixels, with anything beyond that providing diminishing returns, particularly in low light.

If there is an upper limit on the usefulness of increased resolution, the D800 hasn't hit it. We consistently found that the increased resolution provided benefits, specifically in low light—an area where many people predicted the D800 would suffer. The ability to shoot at up to ISO 6400 and downsample to reduce noise naturally, with little penalty, became one of our favorite tricks at dim trade shows like Photokina 2012.

Performance

{{section_header}}{{section.name}}{{/section_header}}

Despite its slow burst speed, the D800 produced excellent lab results, and its real-world performance is even better.

The D800 came through our image quality tests with flying colors. The camera's high resolution sensor excelled in the sharpness test, even taking into account the edge enhancement that's part of Nikon's default JPEG processing. It lagged behind the top Canon bodies for color accuracy, but not by a wide margin. We also found that the camera produced excellent results in our high ISO and dynamic range tests. The D800's 36.3-megapixel, full-frame sensor isn't quite as solid at high ISOs as the Nikon D4, Canon 1D X, or Canon 5D Mark III, but it outpaces all three at the minimum ISO speeds due to its massive resolution.

Of course, test results only tell part of the story. The high resolution sensor allows for considerably more editing leeway, as you simply have more information to work with. We also don't downsample test images to a consistent size for all cameras, which we feel lends an inherent disadvantage to high-resolution models. The D800's shots show more noise when viewed at 100% magnification, but if you're planning on using the images for any normal print size or web display, this noise is hardly visible up to ISO 6400.

There are two specific performance areas that do need addressing, however: shot-to-shot speed and video. The D800's rivals are hardly the fastest cameras in the world—the Canon 5D Mark III only shoots at 6.3fps, while medium format, high-resolution cameras barely top one frame per second—but it's still slower than many would prefer. At 4.14fps, the D800 is just a bit slower than the 5.5fps D600 ($2199.95 body-only), but it leaves Nikon without a new fast full-frame option, except for the $6000 body-only D4 flagship.

In video mode, we also found considerable issues with the D800. While the sensor's output is very sharp and renders motion well, that sharpness comes with an incredible amount of moire. Moire is the ugly coloration you get in patterns as the camera takes 36.3-megapixels of information and samples it down to a normal 1080p video frame. This aliasing is difficult to remove and really mars the D800's otherwise fine video, especially with slanted patterns.

Conclusion

{{section_header}}{{section.name}}{{/section_header}}

An impressive, reliable, professional full-frame body from Nikon

The Nikon D800 is a beast of a camera, an extraordinarily high-resolution land mine, strategically placed amid the abandoned battlefield of the megapixel war. The 36.3-megapixel sensor is easily the D800's greatest asset, making it one of the most flexible, enjoyable cameras we've ever shot with.

If you follow the rumor mill, you know that the D800 was supposed to be a successor to the D700; the D800 was supposed to excel in low light, offer exceptional operational speed, focus quickly, and yet remain light and compact by the standards of bulky full-frame bodies. When it was actually released, though, it was derided by some as an exercise in excess, a body that would necessarily suffer in low light because so many pixels can't fit on a sensor without sacrificing somewhere.

Nikon made some sacrifices with such a high-resolution, but none that impact low-light performance. In addition to our usual round of lab testing, we took the D800 to Photokina in Cologne, Germany for a week. Shooting on dim trade show floors and sparsely lit old town streets, the D800's pixel count turned out to be its greatest advantage. With so much resolution, we were able to edit very effectively throughout the week; softer images could be sharpened up, noise could be reduced, and even cropped images still left plenty of room to work.

Resolution was the biggest advantage in our lab tests, too: The sensor simply provided more data for the processor to work with. This led to some very sharp images, huge dynamic range at low ISO settings, and a very low signal-to-noise ratio—on par with even the Nikon D4 and Canon 1D X. Of course, with all that resolution comes the D800's real limitation: file sizes that limit burst shooting to just 12-15 shots before the camera's buffer fills up.

Otherwise, the D800 is a fantastic camera with exceptional handling and solid design. We'd still caution sports shooters to look elsewhere (like the D4, 1D X, or a used D3), but from dimly-lit streets to trade show floors, the D800 performed as well as we could've hoped. Megapixels have developed a bad reputation over the past few years, but there's no doubt they're a boon to this camera. There are some hitches—spotty left-side autofocus reliability and a distinctly green tint under fluorescent light are very real problems—but the D800 is nevertheless one of the best cameras we've tested, and a real competitor for our 2012 Camera of the Year award.

Science Introduction

{{section_header}}{{section.name}}{{/section_header}}

The Nikon D800 impressed us at nearly every turn, thanks largely to its 36.3-megapixel full-frame sensor. The camera features exceptional dynamic range, excellent noise handling, and generally good-enough color accuracy. We also found that the high-resolution sensor combined with the 24-105mm f/4 lens produced very sharp images, indeed. The only drawbacks came in the form of a pedestrian shot-to-shot speed and increased moire in otherwise sharp video.

Sharpness

{{section_header}}{{section.name}}{{/section_header}}

The D800's images came out looking fantastic, with a remarkable amount of detail brought out in post-processing.

We found when paired with the Nikkor 24-120mm f/4 lens, the D800 was able to produce sharp results across much of the frame and throughout the focal range. The lens only really had issues at the narrowest apertures, where sharpness began to fall off, mostly in the corners. Predictably, we found the lens really excelled at the medium apertures, especially around f/9. We did notice that the camera applies a little bit of extra edge enhancement when shooting JPEGs, though, even in the neutral color mode.

In our real-world shots we observed a little softness at f/4, but the high resolution of the D800 images allowed us to bring a lot of detail back with a simple unsharp mask in Photoshop. We found this was absolutely essential when cropping images down—an obvious strength with so much image information to work with—as images can look quite soft at 100% zoom, straight from the camera.

{{photo_gallery "Science Section 1 Images"}}

Noise Reduction and Detail Loss

{{section_header}}{{section.name}}{{/section_header}}

The D800 handles noise quite well at low ISOs, and the high resolution allows you to downsample to effectively reduce noise for most print sizes.

The Nikon D800's high ISO performance was quite good, with perfectly usable photos up to ISO 3200 for large prints, and ISO 6400 for the web. At Photokina 2012, we shot some of the darker booths handheld with the D800, and even when the auto ISO kicked up to ISO 6400, the shots came back perfectly usable—without noise reduction applied.

The camera's built-in noise reduction system also does well here, as each setting of reduction (low, normal, and high) allows you to use an extra ISO step if need be. We found that noise remained under 2% (our rough threshhold for acceptable quality) all the way to ISO 3200 with no noise reduction, ISO 6400 on low, ISO 12800 on normal, and ISO 25600 on high. There's a level of detail loss associated with each step up, however, so it's still a trade-off.

Since the D800's sensor has a far higher megapixel count than most other cameras in its class, testing this camera was a unique experience for us. This results in smaller photosites on the sensor itself, but the high resolution results in a lower signal to noise ratio overall. It's worth noting here that we do not downsample test images to a consistent size for all cameras. Some sites will resize all images to a uniform size for a more level comparison, which lends an inherent advantage to high-resolution cameras. We think either methodology is valid, but we should note that, at most normal display sizes, the perception of noise in the D800's shots is very minimal.

{{photo_gallery "Science Section 2 Images"}}

Dynamic Range

{{section_header}}{{section.name}}{{/section_header}}

Dynamic range is quite good through most of the ISO range, though by ISO 6400, performance dips noticeably.

The D800's massive 36.3-megapixel image sensor provides it with the ability to capture tones across a very wide range without letting the appearance of noise erode detail. The D800's biggest advantage for dynamic range comes at the low end of the ISO spectrum, where it gains nearly all the advantages of its high-resolution sensor, while the disadvantages of smaller photosites don't yet become an issue.

At ISO 100, we found that the D800 was capable of recording 8.5 stops of dynamic range above an RMS noise threshold of 0.1 (the industry standard places an RMS noise threshold for signal to noise ratio at 1.0, which is a much lower benchmark for quality). That performance held strong until ISO 400, where it fell to 6.91 stops, which is still quite good. From there, performance fell off more dramatically, with just 3.09 stops available at ISO 6400.

{{photo_gallery "Science Section 3 Images"}}

Video Performance

{{section_header}}{{section.name}}{{/section_header}}

Strong showings in most of our performance tests are undone by an ugly aliasing problem.

The Nikon D800's full-frame image sensor produced great motion results in our test videos, but its high resolution caused issues with rolling shutter, moreso when the camera moved than when the subject did. In our still-life motion rig, there's very little ghosting or trailing, and signal interference is practically nil.

The Nikon D800 produced exceptional results in our sharpness test, as we've come to expect from the many test videos shot with the camera here and elsewhere. However, even with the D800's anti-aliasing filter intact (the D800E dispenses with it), moire was still significant in our sharpness footage. It was much better than what we saw with the previous generation of DSLRs, like the Canon 5D Mark II, but it lagged behind the updated 5D Mark III.

Still, the D800 produced a sharper image overall, with nearly 800 lp/ph visible in our sharpness footage. In bright light, we consistently saw 750 lp/ph of sharpness horizontally and vertically, with moire creeping into the shot at higher frequencies.

The D800 was also able to produce a video image with a brightness of 50 IRE using a target illuminated by just 4 lux of light. This puts it among the most sensitive cameras we've tested, though the Canon 5D Mark III, Nikon D4, and Canon 1D X all beat it by a fairly wide margin due to their superior ISO ranges. We should note that this was using the D800 with auto ISO engaged, so it didn't take advantage of the two higher ISO speeds equivalent to 12800 and 25600. Those speeds produce a large amount of grain, but if you need to capture video in less than 4 lux of light, they're available.

{{photo_gallery "Science Section 4 Images"}}

Shot to Shot Speed

{{section_header}}{{section.name}}{{/section_header}}

At just 4fps, this is definitely not the best camera you could pick to capture sporting events.

According to Nikon's specs, the D800 can shoot at up to 4 frames per second. In testing, that was right on the money; the camera captured a five-shot burst at exactly 4.14 frames per second. Again, this puts it among the slower Nikon DSLRs (and the slower DSLRs in the market at large)—right in line with last year's Nikon D5100. Still, 4 fps isn't bad, considering how much information a single D800 exposure contains.

Additionally, we found that the camera was able to capture up to 13 RAW frames in a row before maxing out its internal buffer. When the buffer fills, the camera stops shooting until space is freed up. This happens even when using a CompactFlash card with an SD card for overflow. When shooting full-size JPEGs, however, the burst capacity increased to around 25 shots.

{{photo_gallery "Science Section 5 Images"}}

White Balance

{{section_header}}{{section.name}}{{/section_header}}

Nikon's endemic white balance issues pester the D800.

We found the D800's automatic white balance to be fairly typical, with strong performance under daylight conditions, but with significant issues in tungsten light. The temperature error was roughly 2700 kelvin under tungsten conditions, and just 200 kelvin in daylight.

We also found curiously poor performance under compact white fluorescent lighting, which was off by more than 1800 kelvin. This error produced the "green tint" issue that some D800 users have complained about. Most of these users regard the green tint issue as a problem with the LCD, but we found it in the image files as well. It's almost certainly a flaw in the camera's automatic white balance algorithms.

Custom white balance performance was (unsurprisingly) much better, with tungsten, fluorescent, and daylight conditions all showing temperature errors of less than 150 kelvin. This was the easiest fix when the green tint issue appeared.

{{photo_gallery "Science Section 6 Images"}}

Focus Accuracy

{{section_header}}{{section.name}}{{/section_header}}

The left-point AF issue seldom caused significant trouble in real-world settings.

The D800's 51-point autofocus system was generally quite reliable, both in our testing and in our time using the camera in a variety of real-world conditions. The main issue was focus accuracy on the left side, though supposedly there is a fix available if your camera is affected. You can see an example of this difficulty using a Lens Align Mk. II to analyze the D800's problem here.

We also use a Lens Align Mk. II for our focus test, but we utilize the center focus point to judge focus efficacy in low light. The D800 did quite well in our test, as the center point is much more sensitive than many of the surrounding points. Even in dim 10 lux light in the labs, we found the D800 was able to lock on to subjects with that center point with a degree of accuracy. It won't give you much flexibility with moving subjects, but given the camera's lack of shot-to-shot speed, we don't suggest using it for those purposes.

We should also note that we shot with the D800 for several press events in Cologne, Germany, at Photokina, and for our own purposes too. In thousands of real-world shots, we never once encountered a shot that was clearly ruined by a lack of focus due to the left-side AF issue. While I can't speak for everyone's focusing habits (I tend to shoot with the right AF points or the center point, especially when taking product shots in dim rooms), I will say that the D800's real-world impact may be slightly exaggerated by some. On the other hand, I shot with a f/4 zoom lens; when using a fast-aperture prime, the problem would be more noticeable.

{{photo_gallery "Science Section 7 Images"}}

Other Tests

{{section_header}}{{section.name}}{{/section_header}}

{{photo_gallery "Other Tests Images"}}