Introduction and features
In November 2015 Nikon announced that the D5 was in development to replace the D4S, the company’s flagship SLR aimed at professional photojournalists and sports photographers.
With the Olympic games arriving this summer it was odds-on that we’d see something concrete at CE in January, and sure enough Nikon obliged, revealing its new 20.8MP full-frame model at the show in Las Vegas.
Few at the press conference could claim not to be surprised by the new camera’s top expanded sensitivity setting: ISO3,280,000 or a little over three million (actually, even 280,000 is quite a lot in ISO terms).
This is something Canon hasn’t matched with its new 1DX Mk II, which has a maximum expansion setting of ISO409,600, and it begs the obvious question: does ISO3,280,000 actually produce usable results? More on this later.
While a maximum setting of ISO3,280,000 might attract the headlines, the real reason for this huge hike in sensitivity is the work that Nikon has done to improve image quality and noise control at the more commonly used settings, those within the native range of ISO100-102,400.
According to Dirk Jasper, Product Manager for Nikon Europe, the main target area for improvement with the D5 was the ISO6400 to 12,800 range, as these values are most commonly demanded by professional sports and news photographers.
Another feature that’s incredibly important to the D5’s target audience is the autofocus (AF) system, and Nikon has also upped the ante here, giving the new camera a 153-point system that has 99 cross-type sensors, with a central point that’s sensitive down to -4EV.
Of the 153 points, 55 or 15 are individually selectable – the rest are support points – and the cross-type points extend closer to the edges of the frame than in previous cameras. The system can be set to operate in single-point, 25-, 72- or 153-point dynamic-area AF, 3D tracking, group-area AF or auto-area AF mode.
There’s also a new ASIC (computer) unit that’s dedicated to the AF system to ensure consistently high performance. The AF ‘workflow’ and tracking systems have also been improved, to better keep up with fast-moving subjects.
Of course, a fast AF system needs to be paired with fast continuous shooting capability, and thanks to the EXPEED 5 processing engine the D5 can shoot at up to 12 frames per second (fps) with full autofocusing and metering capability. What’s more, this can be maintained for up to 200 raw files when writing to an XQD card, giving plenty of scope for capturing Usain Bolt’s exploits this summer.
It’s possible to push the continuous shooting rate to 14fps, but when this option is selected focus and exposure are fixed at the start of the shooting sequence. In addition, the mirror stays up from the moment the shutter release is fully depressed, so the viewfinder is blacked out.
The ability to record 4K (3840 x 2160-pixel) video was expected for the D5, but surprisingly it’s only capable of recording for a maximum of three minutes internally. While this will be sufficient in many situations, there will plenty of others where it’ll be nowhere near enough, so the D5 is unlikely to be of interest to serious videographers.
According to Nikon Pro magazine there will be a firmware update that will extend continuous recording to 29 minutes 59 seconds, the same as with Full HD recording, but it still hasn’t materialised.
Some better news about the 4K recording, however, is that it’s possible to employ a native crop so there’s no pixel binning, giving better image quality. There’s also an HDMI output to enable connection to an external monitor.
It’s worth mentioning at this point that there are two varieties of the D5, one that has two CompactFlash (CF) card slots, and one that uses XQD cards. Those wanting the maximum burst depth will need to opt for the QXD model; according to Nikon the burst depth is cut roughly in half when writing to a CF card.
Like the D500, the D5 has a 180,000-pixel RGB metering sensor, which also helps to inform the white balance, scene recognition and AF system.
Build and handling
Anyone with a D4S will feel at home with the D5, as the control layout has been kept the same. The grips on the front and rear of the camera, however, are more ergonomically shaped to make it more comfortable to use for long periods of time. I certainly had no problems holding it for a few hours at a stretch during this test, although a monopod would be beneficial to take some of the weight when shooting with a large lens.
Like the D4S, the D5 feels like it would be safe to use in the toughest of environments, having a full metal body and extensive weather sealing. There’s even a weatherproof cover for the hotshoe, to keep the contacts dry when a flashgun isn’t mounted – I experienced no problems when shooting in rainy conditions during this test.
Although the D5’s control arrangement is very familiar, it’s not absolutely perfect. The mini-joystick sub-selector that’s used to set the AF point when holding the vertical grip, for example, is more awkward to reach with your thumb than the one associated with the horizontal grip.
And, while there are three buttons around the horizontal shutter release, there’s only one by the vertical button. This means you have to adapt the way you control the camera depending on the orientation you’re shooting in – it would be better if the vertical controls were the same as the horizontal grips, although this would require some remodelling of the memory card port area.
While the D5’s 3.2-inch 2,359,000-dot screen is touch-sensitive, it doesn’t make much use of the touch control – it’s only for scrolling through images and zooming in to check sharpness. It works intuitively with taps, swipes and pinch-zooming and responds promptly, but it’s not possible to navigate the menu, make settings selections or set the AF point by touching the screen – it would be helpful if some of the settings displayed when pressing the Info button could be selected and adjusted with a tap.
As I’ve suggested with other Nikon SLRs, it would also be helpful to combine the functions of the Info and ‘i’ button to streamline some of the control. As it stands, pressing the i button reveals a list of eight features that can be selected for adjustment; the list seems somewhat random, and contains items such as Color space and Custom control assignment, which are unlikely to be required on a frequent basis.
Meanwhile, pressing the Info button activates a screen that looks like the Quick Menu of many other cameras, displaying an extensive collection of settings (including some that appear in the i button list), but none of these can be selected for adjustment.
Not surprisingly given the high dot count of the screen, images look great on it, and it’s easy to check sharpness. However, at the default setting I found it often made captured images look brighter than they are, so it’s a good idea to keep an eye on the histogram view, and consider turning down the screen’s illumination.
As it has ‘just’ 20 million pixels on its sensor the D5 can’t compete with the 36MP D810 for detail resolution, but it compares very well with the 24MP D750, matching or beating it throughout the sensitivity range in our resolution tests. It also beats the 16MP D4S up to ISO204,800, after which the older camera pulls ahead – this is deep into the D4S’s expansion range, and one stop into the D5’s.
Significantly, in real-world shooting the D5 produces much better-looking images at ISO409,600 – the D4S’s maximum extension value. The images are still very noisy, and I wouldn’t recommend using this value routinely, but there’s less banding visible; it could be a useful option in extreme circumstances.
Moving up to the D5’s maximum sensitivity setting (ISO3,280,000) results in a significant drop in image quality. In the low lighting conditions that demand such a high setting there’s little detail in either the raw or JPEG files. With no noise reduction applied, raw files have lots of chroma noise (coloured speckling) and a check pattern of banding visible at normal viewing sizes.
Simultaneously captured JPEGs are much softer-looking, as if a watercolour filter has been applied, and there’s magenta banding. There could be a use for this setting for surveillance work, when it may be necessary to record the location of large objects, but it’s unlikely to be used often. Many details are rendered unrecognisable, although some outlines are visible.
The lower high expansion settings produce better results, and could be of use for reporting when image quality isn’t the prime concern, but as a rule I’d recommend keeping below ISO102,400.
The D4S has a strong reputation for autofocus system performance, and the D5 takes things further, not least with a 102-point increase in the number of AF points. When the active AF point is over the subject the D5 does an excellent job of keeping it sharp in single-point AF mode. However, the 25- and 72-point dynamic-area AF options increase the chances of an active AF point being over the subject, and increase the hit rate with subjects that move around the frame as well as changing distance to the camera.
The 153-point and Auto-area AF modes can be helpful, but in busy surroundings there’s an increased chance of the camera latching on to the wrong target – especially in low light and/or low contrast situations.
When photographing at a low-light go-karting track I was acutely aware of when I had the active AF point/area over the subject – when this was the case the camera did an excellent job of keeping the subject sharp. The 3D tracking option can be useful, but it relies on a strong colour contrast between the subject and the background, if this isn’t the case it tends to get distracted, and jump to other areas of the frame.
In other respects the D5 performed well; the Matrix metering system is fairly easily influenced by very dark or bright sections in the scene, but it’s usually predictable. In the high-contrast conditions of the Moscow State Circus, for example the dark surroundings tricked the Matrix system into overexposing the brightly illuminated subject; centre-weighted metering was a better choice.
Colour-wise, the D5 tends towards quite subtle tones in the natural environment when using the Standard Picture Control mode, which is likely to appeal to photographers who are used to processing files. In other instances, with bolder colours, images look more vibrant and punchy – our lab tests indicate that the camera tends towards higher saturation, which is often more attractive.
The video produced by the D5 is of high quality, with good colour and plenty of detail.
Lab tests: Resolution
We chose two rival cameras for the Nikon D5 to see how it measured up in our lab tests: the Nikon D4S and the Canon EOS-1D X Mk II.
We’ve carried out lab tests on the Nikon D5 across its full ISO range for resolution, noise (including signal to noise ratio) and dynamic range. We test the JPEGs shot by the camera, but we also check the performance with raw files. Most enthusiasts and pros prefer to shoot raw, and the results can often be quite different.
Nikon D5 resolution charts
We test camera resolution using an industry-standard ISO test chart that allows precise visual comparisons. This gives us numerical values for resolution in line widths/picture height, and you can see how the Nikon D5 compares with its rivals in the charts below.
JPEG resolution analysis: The D5’s JPEGs either match or beat comparable files from the D4S and Canon 1DX Mk II for resolution. It particularly impresses at the highest sensitivity values – though it’s worth remembering that these measurements are made with a chart that is in good light.
Raw (converted to TIFF) resolution analysis: The D5’s raw files (after conversion to TIFF) follow the same pattern as the JPEGs, indicating that it’s a step up from the D4S for detail resolution, and able to beat the 1DX Mk II at many sensitivities.
Lab tests: Dynamic range
Dynamic range is a measure of the range of tones the sensor can capture. Cameras with low dynamic range will often show ‘blown’ highlights or blocked-in shadows. This test is carried out in controlled conditions using DxO hardware and analysis tools.
Dynamic range is measured in exposure values (EV). The higher the number the wider the range of brightness levels the camera can capture. This falls off with increasing ISO settings because the camera is having to amplify a weaker signal. Raw files capture a higher dynamic range because the image data is unprocessed.
Nikon D5 dynamic range charts
JPEG dynamic range analysis: With the exception of the upper values the D5 performs consistently for much of its sensitivity range, producing images with a wide range of tones.
Raw (converted to TIFF) dynamic range analysis: While the D5’s raw file dynamic range is good, it is beaten by the D4S and 1DX Mark II indicating that these cameras capture a broader range of tones and that images stand-up to a bit more post-capture adjustment. DXO Analyzer was unable to analyse the raw files taken at the D5’s upper sensitivity values.
Lab tests: Signal to noise ratio
This is a test of the camera’s noise levels. The higher the signal to noise ratio, the greater the difference in strength between the real image data and random background noise, so the ‘cleaner’ the image will look. The higher the signal to noise ratio, the better.
Nikon D5 signal to noise ratio charts
JPEG signal to noise ratio analysis: Up to around ISO1600 the D5 lags a little behind the D4S, but above that figure it starts to take the lead, indicating that the higher sensitivity images are cleaner. It’s not until ISO25,600 that the D5 overtakes the 1DX Mk II, but the Canon camera doesn’t fall far behind, and produces the best results at ISO409,600.
Raw (converted to TIFF) signal to noise ratio analysis: Interestingly, the raw files (after conversion to Tiff) don’t compare quite so well to those from the D4S, perhaps indicating the downside of increasing the pixel count, but also suggesting the improvements that have been made with in-camera JPEG processing. The Canon 1DX Mk II beats the competing Nikon cameras at all but the lowest sensitivity values. DXO Analyzer was unable to analyse the raw files taken at the D5’s upper sensitivity values.
Sample Nikon D5 ISO test results
The signal to noise ratio charts use laboratory test equipment, but we also shoot a real-world scene to get a visual indication of the camera’s noise levels across the ISO range. The right side of the scene is darkened deliberately because this makes noise more obvious.
ISO 200: Click here for a full-size version
ISO 6400: Click here for a full-size version
The Nikon D5 is designed for use by professional photographers who need a fast, responsive camera that can shoot in low light and survive the odd knock or splash of water. With its maximum continuous shooting rate of 12fps (when the autofocus and metering systems are engaged), super-quick autofocusing, huge sensitivity range and tank-like build, it certainly delivers in all areas. But does it offer the Nikon-using pro enough to justify upgrading from a D4S?
Increasing the pixel count of the D5 by 25% in comparison with the 16MP D4S enables the new camera to capture more detail than the model it replaces. And although our lab tests indicate that this has a slightly negative impact upon the dynamic range and the signal to noise ratio in raw files, high sensitivity images look very slightly better.
That said, I’d still recommend steering clear of values above ISO102,400, and certainly above ISO409,600 or higher unless it’s imperative to get an image and quality isn’t a major concern.
The D5 has the type of build we expect from a professional-level camera, and while the control arrangement isn’t perfect, existing D4S users will find it easy to swap between models. The touchscreen is a useful addition when reviewing images, but it would have been nice if Nikon had made a bit more use of this feature.
Even in low light the D5’s autofocusing system is extremely capable, and it’s possible to get long sequences of sharp images of fast-moving subjects.
As well as providing more room for a larger battery, the purpose of the vertical grip on twin-gripped cameras is to make them easier to hold and operate when shooting portrait-format images. The D5’s vertical grip goes some way to achieving this, but it’s not as successful as the Canon 1DX Mk II’s implementation.
The AF controller (aka sub-selector) isn’t in the same location relative to the grip, so as you swap from horizontal to upright shooting you find yourself reaching towards the wrong place. This is something D4S photographers will be well used to, and which most will have adapted to, but it’s not ideal. Nikon could also do with rethinking some of the interface to make key features easier to find and adjust quickly.
The counter argument to these handling issues is that the D5 has been kept the same to make it easier for working photographers to upgrade from the D4S, or keep their D4S as a second body to the D5. But if that’s the case, does that mean we can never have any handling improvements?
The D5’s limited 4K recording capability is disappointing, and it rules the camera out for many photographers wishing to shoot both stills and video professionally. This may change if the promised firmware upgrade arrives soon.
The Nikon D5 is a great camera and a worthwhile upgrade to the D4S, but there are a few niggles that if addressed could make it an absolute belter. The top sensitivity expansion settings are headline-grabbing and largely useless to most photographers, but the image quality in the native sensitivity range is high, provided you keep an eye on the exposure.
Autofocusing is blisteringly fast, and able to keep up with fast-moving subjects and a continuous shooting speed of 12 frames per second, making the D5 a great choice for professional photographers looking to shoot this summer’s Olympic games – but a few may look enviously at their Canon-toting colleagues, who have better video capability and more streamlined handling.
Powered by WPeMatico