January 2012. Nikon launches the D4. Time for a review of its product strategy.

With the launch of Nikon's new flagship D4 and the rumored specs of the upcoming D800 – supposedly to be launched on February 7 – it seems like a good time to try and take a look at what may be on their (marketing) mind, especially because everybody has already extensively reviewed the D4's specs, while no hands-on tests are yet to be found.

During the last decade or so, Nikon have had more than their fair share of “game-changers”, like the D1 (first professional digital camera for under 10K), the D700 and D3S (still DxOMark top 1 and 2 in high ISO), the D3X, third in DxOMark's overall score, only behind two Phase One digital backs, and the D90 (first DSLR with video), amongst others.

Still, they have also been late in many other ways.

Nikon was extremely late in going from CCD to CMOS and moving from DX to full-frame (FX), which is why Canon lead the clean high ISO race for the best part of the 2000's, until Nikon changed the game with the arrival of the D3 (2007), D700 (2008) and more decisively with the D3S in 2009.
Although they were the first to launch a DSLR with video capabilities (D90), its implementation was rather rudimentary, which is why this lead was soon lost to Canon, which has been selling the hell out of its EOS 5D Mk II ever since.

The new D4's rather modest 16,2 Mp. appears to be yet another indication that superb low light, high ISO performance is still only possible with relatively low pixel densities and large pixel pitches; 7,3 mµ in the case of the D4 and a whopping 8,5 mµ in the case of the D3S and D700. Compare that to the 5,5 mµ of the D300S, the 4,8 mµ of the D7000 or the 3,4 mµ of the Nikon 1 system.
The pixel pitch of the D800 – if and when the rumored specs are correct – would be 4,9 mµ in the case it indeed features a 36 Mp. sensor.

Of course there are many other factors that influence good high ISO performance.
Both Nikon (D3S) and Sony (BSI-chips) claim to achieve considerable improvements in SNR (signal to noise ratio) through the use of an optimized MLA (micro-lens array) and better color filters, while Nikon's improved EXPEED-3 processing engine would explain why their cameras generally still perform better than their Sony brethren, even when using the same sensors.
Therefor, a 36 Mp. D800 – even with its much smaller pixel pitch in comparison to other FX Nikons – ought to still score better than the D7000, in spite of their similarity in pixel pitch and -density.
This is not at all an unreasonable assumption to make, considering that the D5100 – launched in 2011 – already outscores the D7000 (2010) in DxOMark Low Light ISO, even though not in overall performance.

The D800's rumored standard ISO range is 100-6.400, expandable to 25.600 – exactly the same as the D7000 and D5100 – which would give us a camera with a resolution akin to mid-format, with at least the high ISO performance of a current top of the pack DX camera, pulling it more on less on par with the Phase One P40 Plus (33x44 mm - 40,1 Mp. - 5,85 mµ - U$ 14.000).
The D4, on the other hand, gives us one more stop of high ISO compared with the D3S (204.800 vs. 102.400) which would keep it in the race against the upcoming 18,4 Mp. Canon EOS 1D X.
One would expect its – considerable – improvements in video performance also to be included in the D800, which would give Nikon – in theory – not one, but two distinctly different DSLR's to threaten the hegemony of the EOS 5D Mk. II in video-land.

There would also be very little reason to assume that Nikon is willing to give up its lead in low-light, high ISO performance, which is why the D4 will probably outperform the D3S, in spite of its 16% higher resolution and corresponding smaller pixel pitch, resulting in the D4 replacing the D3S as the Nº 1 camera in low light, high ISO performance in DxOMark.

Meanwhile, with 22% more resolution than the D3X, the D800 clearly aims at taking a top spot in both the Portrait and Landscape categories, where the D3X is currently already 4th and 3rd respectively.
Slotting out the D700 in the high ISO category (currently 2nd), would leave the D3-D4 series dominating the sports & journalism section, while the D800 should be taking a shot at a top rank for studio-, portrait- and landscape work, including ahead of some very respectable mid-format cameras by Phase One, Hasselblad, Pentax and the likes.

Unlike the D700, being a “semi-professional”, comparatively cheap and cannibalizing clone of the flagship, it seems that Nikon is looking to position the D800 (± U$ 3.000) and especially the D800E without AA filter ($ 3.300) as economical alternatives for mid-format cameras, which is not THAT stupid, considering that the cheapest mid-format – the Pentax 645D – streets for slightly under 10K, while cameras with interchangeable digital backs start counting at around 15.

Not a bad bet – say I – even if people who were waiting for a D700 with video capabilities will most likely be disappointed by Nikon's reorientation of it's top (full-frame) product lines.
Considering that a FX 36 Mp. sensor is at the very limits or beyond of what even the best standard production lenses can resolve – as this Luminous Landscape article explains – deciding if the D800 is worth the money is a tough decision that should not only consider the theoretical, but most certainly the practical: do you have the glass and, more importantly, the skill to take advantage of it?

See also: Nikon and Adobe add support for the D4, D00 and D800 E.