It certainly took long enough, but my D800 finally arrived today! I haven’t really gotten a chance to use it yet – but I don’t think my opinion matters much anyways. When I pondered whether the D90 or D7000 would be a better choice for other non-photographers that like me were primarily going to take pictures and capture memories as opposed to creating art, that might have been useful. On the other hand, the D800 seems awesome, but it is clearly overkill for the non-photographer, and if you’re going to spend a big chunk of money on a camera followed by a potentially bigger chunk on full frame lenses, then you probably should read through the 25-page professional reviews and make an informed decision.
That being said, highly preliminary use has been pretty encouraging – with some of the anticipated drawbacks:
- I’m comfortable raising my Auto-ISO limit to 6400. Per-pixel noise indeed beats the D7000 by perhaps a half-stop at the same ISO setting… and at any given viewing size, there’s a little over twice as many pixels in a shot from the D800 vs. D7000. In practice, that means in low light situations, I’ll have twice as much shutter speed (or more depth of field) than before, and since the kids seem to increase their top speed faster than Moore’s law, that really helps.
- A DX crop looks better than a DX camera. The D7000 was generally considered to have the best image quality in a DX form factor (though the new entry-level D3200 has a new 24MP sensor that hasn’t been fully evaluated yet). With any lens, DX or FX, the D800 indeed seems to produce a slightly better image when cropped to a DX level. This is actually pretty amazing; with the 24-70, I can get full-frame quality in the 24mm to 70mm range, but I can crop all the way to an equivalent of 105mm and still have slightly better results than the same lens on the D7000 with no cropping.
- Great viewfinder! I forgot how nice the big viewfinders on the D700 and D3 were – but this is instantly apparent the first time you raise the D800 to your eye. It’s not better than those cameras, but is nice if you’ve used something smaller for a while.
- Not too bulky. My earlier speculation based on weight stats (900g vs. 700g) holds up in reality; it is a little bulkier, but not by much, especially when you consider total weight with a lens attached. It’s a far cry from the D3, which was the last camera besides the D7000 that I used in any depth before this one.
- Files are huge! RAW files at default settings are 50MB a piece, and slow Lightroom (and other software) significantly. JPEGs at the Quality 100 setting I was using (since I backup the JPEGs; RAW files are only backed up within my home) were 24MB on something that was a little cropped. I’ve changed JPEG quality to 90 to compensate; I can’t tell the difference visually, and it cut at least the file size in my experiment by 50% to 16MB.
- No ML-L3.The Nikon ML-L3 was a cheap ($18) remote trigger for the D90, D7000, and many other models. I use it on tripod shots like the above to avoid effects of camera shake. For the D800 and up, Nikon expects you to buy a $125 non-wireless remote trigger instead.
It’s not really fair comparing the D800 vs. D7000, since one costs almost three times as much as the other; almost everyone would be better getting the cheaper D7000 since you can get a whole range of top-tier lenses or accessories with the $2000 price difference, and except in very specific conditions, that will have more positive impact on your results.
I really haven’t had time or opportunity to take real pictures with the camera yet, though here’s one shot of Leo in the worst possible conditions:
What do I mean by “worst possible conditions”? Our living room is the most dimly lit room in our house, and this is 8pm on a cloudy evening so there was very little natural light. The above shot is at 50mm, f/1.4, ISO 6400, 1/125th. The 50mm f/1.4D isn’t a great lens wide open, but it’s the only f/1.4 lens I have. The 24-70/2.8 would have dropped the shutter speed to a mere 1/30, too slow to handle either camera shake or kid shake. I used minimal noise reduction (10 in LR4), and while there’s plenty of pixel-level noise, the end result is still usable even with terrible light (quality and quantity).
For comparison, the same conditions with my Galaxy Nexus:
I’ve talked about the magnitude of the difference that exists between smartphones and cameras, but low light conditions are really where those differences become visible. The D800 is shooting at 10 times higher ISO yet it looks 10 times better, even on static things like the piano which aren’t affected by the vastly lower shutter speed that the Galaxy Nexus was using. That said, what you see in the Galaxy Nexus picture is far closer to what things looked like to the eye – very dim, with the last light of the day in the background. The D800 absolutely has better vision than a human being!