Here I capture a few of my thoughts about the D800 / D800E feature set.
I have a D800E and D800 coming. I am buying the D800E, getting a loaner D800. I will be comparing both, and the Nikon D4 also.
Nikon appears to have produced a HOME RUN, especially since Canon has nothing near it (yet). No doubt the D800 / D800E will have some disappointments and limitations, but it looks about as appealing to me as any Nikon DSLR yet.
Will it be as good in image quality as the Nikon D3x? Well, 2+ years of sensor technology development has to count for something. Possibly better, possibly not quite as good. But even it it were close to the D3x on a per-pixel basis, having 50% more pixels is an intriguing tradeoff for some shooting.
Sampling the image at 36 megapixels opens up new horizons in photography, new creative possibilities, increased realism, the ability to crop, to downsample, etc. I’ll suggest that 100-200 megapixels would be ideal for the sampling frequency of a DSLR, which might mean finished images in the 50-60 megapixel range. So we are not done yet.
Let the megapixel haters stew in their own mental vomit.
How will Canon respond? Will a Canon 5D Mark III appear with similar megapixels? Or will Canon claim the smaller-is-better approach? I predict that failure to match the D800 resolution will grow Nikon’s market share substantially.
Note that by going Nikon, one can buy Zeiss ZF.2 lenses, and these can ALSO be used on Canon with an adapter. The reverse is not possible.
“But I only want 18MP JPEGs in sRGB”
Easy: shoot in RAW, batch process to beautiful artifact-free 18MP JPEGs with proper white balance and sharpening and low compression, then toss the RAW files. No big deal. And a big deal: superior quality including the option to rescue an image or use a wide-gamut space when sRGB or AdobeRGB would flatten the color response (increasingly common).
Vs the Nikon D4
For those who must have ultra low-light capability, the Nikon D800E + Nikon D4 seem to me to be a tantalizing combination.
Based on Nikon’s past history (e.g., D3x / D3s), and the Sony NEX-7 image quality (even smaller photosites), I predict that the 2-3 years of sensor technology improvements are going to produce stunning image quality at ISO 100 - 400, along with very good image quality thereafter. This is very exciting and useful, given the 36MP resolution.
Setting aside the cost of the D800, it also means that a cropped sensor DX-format camera is pointless for the purposes of getting closer to a distant subject (e.g. wildlife), since the D800 can be cropped down to its DX area for 16MP images. Thus the D800 represents a best of all worlds scenario.
I am disappointed here. The 921K pixel screen is claimed to have some improvements in viewing quality, but I was hoping that the 36MP sensor would be matched by a 1.5 or 2+ million pixel LCD.
I hope that the ambient light sensor for the LCD does not go haywire when using a focusing loupe.
In terms of per-pixel quality, Nikon’s discussion of 14-bit A/D conversion and a 16-bit internal processing pipeline is highly encouraging. Which suggests to me that with even halfway decent lighting and the very best lenses, we might be in for a jaw-dropping camera. And that with most lenses, the camera will be so good that only the failure to use a better lens will keep the photographer thinking that the optical quality is good. Eyes might be opened with the right proper select few lenses.
I predict that most Nikon zooms and some primes will not be up to snuff over the whole frame, but will provide very acceptable results in the central 1/2 to 2/3 area, which is where portraits and similar shots need the sharpness. I doubt very much that landscape and similar shots which require whole-frame sharpness will realize the potential of the D800 using most of the Nikon lens lineup, at least most of the zoom lineup. The D800 / D800E will even flush out the few weaker offerings in the Zeiss lineup.
One reader asked me if I would be reviewing the Nikon 28-300 zoom! It almost felt as if my 4 year old were asking me about Santa Claus or the Tooth Fairy. While zooms are much better than they were even 5 years ago, one’s image quality standards have to be rock-bottom to posit that a 28-300 would be a good lens on a 36MP camera. I shot the 28-300 briefly on the 12MP Nikon D3s, and it was an underperformer there, and that’s 12MP, not 36MP. Shoot a top-grade lens, and your expectations change, forever.
Reader Norman M writes:
Here's hoping to some explorations using a tilt-shift lens in your upcoming review! I have been seriously considering moving to this sort of setup for my photography as I mainly shoot landscapes, and think that having the advantages of lens movements can really increase one's capabilities in the field.
It is true that tilt can be very helpful in evading the depth of field issue, so long as the subject matter has a planar or semi-planar outline, so that a lens tilt can align the zone of sharp focus along that plane. With landscape, I frequently find tilt less than useful because of oddball subject shape, but when appropriate, it’s a powerful tool.
Face tracking might actually be useful. I don’t like most silly new features, but I am going to keep an open mind on this one.
There is a first time for everything I suppose. Like clockwork, Nikon claims improved metering once again. Except that I resort to manual frequently with every past Nikon body, because the AF system is too unreliable for any of the interesting scenes that I shoot.
I don’t believe in the Tooth Fairy or Santa Claus. But I might take a fresh look if Nikon fixes the AF accuracy issues with my D3x and D3s with f/1.4 lenses.
I keep thinking video might be useful for me. But I’m intimidated by video. Really. It’s a whole different art form. Still, the size and weight of the camera mean that it might yet prove to be the first video camera I really use. Of course I would use my Zeiss lenses, because Nikon AF lenses for manual focus are an exercise in frustration.
Too bad the D800 cannot shoot 4K video.
I do hope that Nikon has fixed the white-out bug in the D3x and the bang-bang mirror slap problem and has implemented Canon-style “Mode 2” exposure, and also has not gone and introduced some fundamental flaw.
Jeff L writes on Live View:
Your comment is for me the most important question I want answered before I consider purchasing a D800E or a D4. I have used the D3X long enough to learn that the mirror bang-bang in Live View mode is an incredible detriment to a macro shooter such as myself. Fortunately I still have a Canon body around I can employ when I need to eliminate the bang bang.
The answer will be apparent as soon as you have any of the new Nikon bodies in your hands. So please when you do the initial commentary you always do when you first received any of the cameras PLEASE mention this in your blog ASAP.
If Nikon got this wrong I will be really pissed. There is just no excuse for the current implementation.
Peter K in Amsterdam writes:
I have heard from the Nikon guys that instead of one , they now have 3 different motors running the mirror- shutter-aperture mechanism. ( a first change since 1984 the F301 !)
- Now if the mirror is up it does not use any power to stay there.
- No more slapping mirror during live view.
- Micro aperture control up on 1/8 stop ( also during video).
- Also very good and new because the shutter was very accurate but the aperture al lot less.
- Very new quiet mode-- mirror stays up only the shutter moves
So for sure things has changed. Also I think the new light metering will do a far better job, but we will have to see it in practice...
My review in DAP will make all such points right away. so readers should study my review carefully, and sign up for the mailing list to be notified of new additions. The blog is a 2nd priority and when heavily into a demanding review, all I might have time for is to note that I have posted some new pages.
Is it too much to hope that self timer mode could be set to mirror up immediately, so that image quality is not destroyed by vibration when using the self timer with an unstable setup or long lens?
Being able to set an exposure longer than 30 seconds without resorting to a programmable remote release. At least let us have 1/2/4/8 minutes Nikon. Please.
Useless for Mac users as yet. But what could Nikon do.
I thought were were finally done with the dual card slots taking incompatible cards. I suppose one can just ignore one of the slots. But now if you also carry a Nikon D4, that high speed XQD card is useless in a D800/D800E, leading to one more planning complication for those of us carrying a system with both cameras.
Somewhat useful yes. Accurate, probably not. The existing virtual horizon feature on the D3x / D3s is not accurate to more than ~1°, so buildings visibly tilt, something I’ve seen over and over with my Nikon pro bodies.