Nikon D800 is outed with pics and specs. 36 MPX

[tsjanik]

......As far as lack of detail goes, does it also apply to this image?

http://chsvimg.nikon.com/lineup/dslr/d800/img/sample02/img_06_l.jpg

.......

........I don't believe the camera is actually that bad and would like to see samples that use proper technique (tripod, proper processing, etc). Some of the images have obvious halos around things like branches as well. Here is an example: http://chsvimg.nikon.com/lineup/dslr/d800/img/sample02/img_06_l.jpg

Nikon should take those down and update them with proper files because this is quite embarrassing for a 36mp sensor.

Same photos, different conclusions.  Interesting.

[BernardLanguillier]

Same photos, different conclusions.  Interesting.

I do agree that some of the branches do have a halo around them, but still think that the resolving of fine details is excellent.

The halo is most present in the strongly back lit areas left of the center of the image, you have a lot less to the right. I feel that the way to sharpen these AA filter less images is probably different.

I would be interesting to see how well Capture one pro is going to perform with D800E files, hoping that Phaseone doesn't consider the D800 as a competitor like they do with the Pentax 645.

Cheers,
Bernard

[DaveCurtis]

There seems to be a lack of micro contrast in the images and one image seems to show asymmetric sharpness. Rather blury on the left.

It's difficult to say whether the lack of micro contrast is the lens used or the raw development. The images show good resolution in the central area.

[Ray]

Indeed, the 14-24 f2.8 will have zero problem with the D800. It was designed from scratch to be able to deal with this level of resolution down to the very corners. This lens is without doubt one of the strongest assets of the Nikon system and Canon has not been able to replicate it yet more than 4 years after its release. Many EOS users are going through the trouble of using an F mount adapter so as to be able to shoot with the 14-24 f2.8.

The only wide angle lens coming close is the Zeiss 21mm f2.8. For the rest the 14-24 f2.8 is in a class of its own. It does cover the 35 mm format far better than the fixed focal length Hassy and Phaseone 28mm f4.0 cover their respective formats. I don't know about the Pentax and Leica equivalents that are said to be excellent, but of course are far from being as wide and are fixed focal lenses.

Cheers,
Bernard

I think I've mentioned a few times, the quality of this Nikkor 14-24/2.8 is the main reason why I bought a Nikon D700, 2 or 3 years, ago despite being a long-time Canon user. I later bought a D7000 with 24-120/F4 zoom, so these two cameras and lenses are now what I use for most of my photography. However, when I need a long telephoto lens, I'll still pick up my Canon 50D with 100-400/F5.6 IS zoom.

For me, the D800E will be a good replacement for both of these Nikon cameras. Because the pixel density of the D800 is close to that of the D7000, the D7000 will have no crop-factor advantage and will therefore be basically redundant, apart from those occasions when it's advantageous to carry two cameras with lenses attached in order to capture the fleeting moment which doesn't wait for one to change lenses.

[larkis]

I do agree that some of the branches do have a halo around them, but still think that the resolving of fine details is excellent.

The halo is most present in the strongly back lit areas left of the center of the image, you have a lot less to the right. I feel that the way to sharpen these AA filter less images is probably different.

I would be interesting to see how well Capture one pro is going to perform with D800E files, hoping that Phaseone doesn't consider the D800 as a competitor like they do with the Pentax 645.

Cheers,
Bernard

At the risk of starting a flame war, I put together this image that compares the Pentax 645D vs the samples Nikon posted. http://blog.dominik.ca/2012/02/07/nikon-d800e-vs-pentax-645d/

It seems that Lloyd Chambers observed the same issues I have when looking at the Nikon samples. http://diglloyd.com/blog/2012/20120207_8-NikonD800-Samples.html

I'm also positive that I can get more detail from the Pentax by using a better tripod and more careful technique. The longer exposures on the Pentax frames should have put it at an disadvantage against the Nikon.

[BernardLanguillier]

At the risk of starting a flame war, I put together this image that compares the Pentax 645D vs the samples Nikon posted. http://blog.dominik.ca/2012/02/07/nikon-d800e-vs-pentax-645d/

It seems that Lloyd Chambers observed the same issues I have when looking at the Nikon samples. http://diglloyd.com/blog/2012/20120207_8-NikonD800-Samples.html

I'm also positive that I can get more detail from the Pentax by using a better tripod and more careful technique. The longer exposures on the Pentax frames should have put it at an disadvantage against the Nikon.

The Pentax images are for such much better.

Cheers,
Bernard

[torger]

I think it is a bit early to come to any conclusions based on Nikon JPEG samples, I would not be surprised if less-than-perfect technique was employed during shooting, and JPEG compression reduces microcontrast.

However it would be a total disaster for the (lower res) medium format cameras if they do not show an significant edge, costing many times more and being less versatile. Sony's sensors have shown to perform very well, so I'm sure medium format at similar resolution won't have much of an edge concerning the sensor. The value of having larger area is often greatly exaggerated - at base ISO photon shot noise is not a problem for 24x36, the difference up to say 44x33 (typical size of entry level medium format) is negligible.

The advantage of medium format will instead be on the lens side I think, where you can have larger pixels for the same resolution and thus stress the resolving power of the lenses less.

At 36 megapixels it will be very evident which lenses that are sharp and which are not. Which are sharp and which are not will depend very much on which optical designs that are possible for a specific focal length, plus how well executed they are of course. Some focal lengths will perform very well, while others will not. 50, 85 mm will be good, 70-200 zoom will be good. Ultra wide lenses will be poor. I'm more familiar with Canon lens performance though, Nikon are better than Canon on some focal lengths and vice versa.

[BernardLanguillier]

At 36 megapixels it will be very evident which lenses that are sharp and which are not. Which are sharp and which are not will depend very much on which optical designs that are possible for a specific focal length, plus how well executed they are of course. Some focal lengths will perform very well, while others will not. 50, 85 mm will be good, 70-200 zoom will be good. Ultra wide lenses will be poor.

Indeed. I am not too worried, the Zeiss ZF 100 f2.0 and 50 f2.0 I have been using should hopefully be up to the task. 

Cheers,
Bernard

[Chairman Bill]

I seem to remember less than stellar results in Nikon's photos at the launch of the D700. I was less than impressed. And then people started shooting with the camera, et voila! decent images. I'll defer judgement until I've seen what we get from other sources. Personally, I'm just hoping the market now gets (relatively) flooded with D3x bodies, & I can pick one up cheap(ish).

[tsjanik]

At the risk of starting a flame war, I put together this image that compares the Pentax 645D vs the samples Nikon posted. http://blog.dominik.ca/2012/02/07/nikon-d800e-vs-pentax-645d/

It seems that Lloyd Chambers observed the same issues I have when looking at the Nikon samples. http://diglloyd.com/blog/2012/20120207_8-NikonD800-Samples.html

I'm also positive that I can get more detail from the Pentax by using a better tripod and more careful technique. The longer exposures on the Pentax frames should have put it at an disadvantage against the Nikon.

Thanks for the comparison.  Your shots also illustrate just how good the 120mm is (and that the new 55 is not quite in the same league).

Tom

[uaiomex]

Very slight traces of banding I see at the left over the curtain and under the Fuji lens. Very clean for 25,600.
here:
http://mansurovs.com/wp-content/uploads/2012/02/Nikon-D800-High-ISO-Sample-4.jpg

[BJL]

[Aside to deejjjaaaa: Chris Sanderson is right that this topic belongs here, not in the thread he closed.]

The question is why the D800E still has the two somewhat pricy slabs of special glass that do OLPF filtering, but oriented so that the filtering effect does not happen.

My guess is based on this fact: if they were removed, leaving a thinner layer of glass in front of the sensor and more air, the optical path length (and light travel time) would be shorter due to the lower refractive index of air, and this would move the focal point. Even a flat slab of glass in the path of light moves the focus. Also, the optical elements sandwiched between the two parts of the OLPF (I guess IR cut filter, as Imaging Resource also does) would have to move closer to the sensor, unless Nikon could put an air gap in there, again changing the optics. So for Nikon to simply leave out the OLPF is the far smaller number of cameras that it is likely to sell in the "more resolution, more aliasing" version E, it would have to either move the viewfinder screen, or the reflex mirror, or the sensor, to restore matching between the optical path to the sensor and the optical path to the VF screen. Maybe all that redesign and a second slightly different fabrication setup would cost more than leaving in the two slabs of special OLPF glass.


Illustration from Nikon via Imaging Resource:

[jhmaw]

This is all very well, but how many photographers have lenses capable of matching the resolution of the camera. Throw out your zooms now. Only a very few Nikon prime lenses will do the job.

Frankly, if you really need that sort of resolution you are probably better off with medium format. I think this camera will appeal primarily to those who are endlessly chasing bigger and bigger numbers, mainly for the sake of it.

As for some of the comments about it hurting Canon, how many people will ditch their system just for these numbers. Again, probably only the number chasers (do those people actually use cameras or just talk about them?).

[BJL]

This is all very well, but how many photographers have lenses capable of matching the resolution of the camera. Throw out your zooms now. Only a very few Nikon prime lenses will do the job.

Frankly, if you really need that sort of resolution you are probably better off with medium format.

Firstly, a $3000 high resolution SLR is clearly only intended for a small group of photographers, who are likely to either own good sharp Nikon lenses, or be willing to buy some. And when you consider DMF as an alternative, note that this $3000 camera leaves about $7,000 or more spare to help buy the needed Nikon lenses compared to even entry-level DMF pricing.

Secondly, it seems that some Nikon zooms are also up to the task, but when you approach MF resolution territory, you might have to accept the "MF lifestyle" of using mostly prime lenses.

Thirdly, lens resolution does not end suddenly at some limit in lines per mm [l/mm], so what happens when sensor resolution increases "too much" is just that the overall resolution goes up less than the sensor alone would indicate, buy you still get more resolution in l/mm or better contrast at equal l/mm, than with the same lens and a lower-resolution sensor. This point can be lost on people who only compare alternatives with different pixel counts at 100% pixels on-screen rather than under equal viewing conditions, meaning viewing equal sized images from equal distance.

[seanesopenko]

And when you consider DMF as an alternative, note that this $3000 camera leaves about $7,000 or more spare to help buy the needed Nikon lenses compared to even entry-level DMF pricing.

Exactly.  $7000 is a lot of money.  Think of all the lenses one could get for $7000?  Or all the travel?  Or even a nice 24 inch printer with cash to spare?

[kers]

This is all very well, but how many photographers have lenses capable of matching the resolution of the camera. Throw out your zooms now. Only a very few Nikon prime lenses will do the job.
Frankly, if you really need that sort of resolution you are probably better off with medium format. I think this camera will appeal primarily to those who are endlessly chasing bigger and bigger numbers, mainly for the sake of it...

I think you are right… for now
My idea is that the sensor itself and camera could be very good.
So then you only need the right glass. ..expensive lenses.

At the moment only the teleprimes are very expensive.. and very good.
They are made because there is a market for them. This market will grow quickly to normal and wide-angle lenses as soon as MDF people are interested in using  35mm cameras. I am sure this interest is there.
Already Nikon and Canon are producing new high quality lenses to feed that need and it is only the start. If not, we have Zeiss et al to fill that gap.

[Christoph C. Feldhaim]

This camera together with Zeiss primes must be a real killer.
I'm curious on the first test shots of this kind.

But for me it also appears clear, that with these extreme resolutions the limits of the format will soonish be reached:
1.) You'll need expensive first class lenses to feed the sensor.
2.) You'll need to keep the F-Stop low enough since diffraction comes in sooner (with a pixel pitch of 4.88µ the  Airydisk will fully cover a sensel at about F 5.0, at F 11.0 four sensels will fully be covered by the Airy disc)
3.) You'll need to be even more consequent in your shooting technique, use tripod, MUP and so on.

If one of 1-3 is not met it is very likely you're just going to produce data junk on your Harddisk which contains 50-75% mush.

[MrSmith]

This is all very well, but how many photographers have lenses capable of matching the resolution of the camera. Throw out your zooms now. Only a very few Nikon prime lenses will do the job.

Frankly, if you really need that sort of resolution you are probably better off with medium format. I think this camera will appeal primarily to those who are endlessly chasing bigger and bigger numbers, mainly for the sake of it.

if you want to hand clients 100meg tiffs (or be able to crop to 60) and shoot in poor light at 400-800asa, shoot handheld and focus accurately then you are not better off with medium format.
there are lenses out there that are more than a match for this kind of sensor. not a nikon user but the zeiss 50mm makro-planar f2 will be on my shopping list when canon bring out something similar, the 17,24 and 90 tse are all very good and the 70-200. the new 24-70 we don't know but i presume it's been updated to have the resolving power. i'm sure nikon have some good lenses? the short zoom is meant to be fantastic.

a pointless upgrade for the Flickr merchants and lurid HDR fans but for working photographers who need the file size i think £2k (vat off) is an absolute bargain.

[mmurph]

You should see what even models with pro makeup look like using MF backs, it ain't pretty at 100%....

Yes, I am going to make my female models shave just 20 minutes before the shoot! 

I pre-ordered the 800E.  I am a lifetime Canon user.  I too hope they announce the 5D3 before the 800E ships.  I have a lot of glass I really like and don't feel like swapping out.  I will especially miss:

* 85 1.2

* 45 tilt/shift

*17 tilt/shift

* 24-70 II just coming out (MTF looks great compared to teh very good version I)

But the 5D2 has been around for 3 years and is still a solid camera. So at least these bodies will stick around for a while and make the switch worth the trouble.  (I only have a 40D right now for pesonal reasons and need a full frame again by April.)

Maybe though Canon is going to announced the 40MP rumoured camera? Those 3 new lenses may be a precursor?

[mmurph]

I think it is worth mentioning that many of the sample shots out there were taken between July and September of 2011.  One linked above was the 14-24 zoom 15mm at f/8.0

Many companies will put an embargo on early images from an alpha camera.  While Nikon is apparently comfortable with those images, firmware and other software tweaks are often underway up to and even past the first shipment date.  

The pixel size is not that much different than that of MFDBs like the IQ180. There is no reason to expect much less than the kind of results that sensor is capable of.  The CMOS should give it a bit of an edge over some of the MF CCD backs, as should the newer design and more up to date software (than some older backs.)

There are a few folks around who were able to tweak their tools and techniques to get great quality out of the otherwise disappointing Kodak 14N.

I was quite happy with the majority of my Canon 1DsII images at 18"x27", 180dpi native without rezzing up.  I sold all of my MF 645 and 6x7 film equipment after I got the 1DsII.  I hope that the 27"x 40" 200-210 dpi images of the 800E can tweak out a bit more quality without the AA filter.  

I print everything at a minimum size of 20"x30".  For larger images I have been using 4x5 film.  I have 2 24" printers and a 44". (Just to answer the "nobody is going to change systems, except those that don't actually make pictures" comments.)