Mother of God - D800 scores 95 DxOMark

[helpful]

I congratulate Nikon on producing a stellar camera in the D800. Now can someone please post sample images taken with both a D800 and a 5DIII, where the D800 made them a better photographer? I have a feeling that I'll be waiting for a long time.

Easy to spot Canon images they are the ones with banding in the shadows & with burnt out highlights  & very poor  reds , I love the Canon red colour its a great red just like to see more shades of red sometimes.

ROFL, I'm almost an old man, but I'm still literally almost rolling on the floor with gales of laughter. You have the greatest sense of humor, or should I say camera sensor for humor. When I got to the "love the Canon red" that really did me in. LOLOLOL.

[dilbert]

...
So the D800 is about the same as the 1ds3 then at low iso ...... big news item 'Nikon catches up with 4 year old Canon tehnology'

No, the 1Ds3 never had DR like the D800 does at low iso.

The 1ds3 outruns the 5DII which is just behind the D800 .....

Are you looking at the same camera?

1Ds3..
Color Depth   24 bits
Dynamic Range  12 Evs
Low-Light ISO   1663 ISO

5D2
Color Depth 23.7
Dynamic Range 11.9
Low-Light ISO   1815 ISO

D800
Color depth 25.3
Dyamic Range 14.4
Low-Light ISO  2853 ISO

Perhaps the most damning of all is the DR - as can be seen in the graph below.

In some of the other graphs, the D800 is close to a full stop better than either the 1Ds3 or 5D2.

That means the D800 at ISO 200 is generating images as good as the 5D2 at ISO 100.

This isn't "just a little bit better", it is significantly better and what we expected/wanted from the 5D3.

[dr croubie]

The biggest problem with most of the canon sensors is that the performance 'plateaus' at lower ISO, like iso100 isn't much better than iso200 or iso400. The D3X pretty much went in a straight line, and the D800 does too (just a bit better than the D3X).

Something tells me that they haven't fixed it though, because canon marketers would be all "4-stop improvement in image quality at low iso". They've told us high-iso is a stop or two better than 5D2, and i'd believe that. What i'd like to see is better DR etc at iso100. D800 can obviously do it, i hope the 5D3 can too...

[dilbert]

The biggest problem with most of the canon sensors is that the performance 'plateaus' at lower ISO, like iso100 isn't much better than iso200 or iso400. The D3X pretty much went in a straight line, and the D800 does too (just a bit better than the D3X).

Something tells me that they haven't fixed it though, because canon marketers would be all "4-stop improvement in image quality at low iso". They've told us high-iso is a stop or two better than 5D2, and i'd believe that.

That is for in-camera produced JPEG files only, which means the software in the camera is better, not the hardware.

[dr croubie]

That is for in-camera produced JPEG files only, which means the software in the camera is better, not the hardware.

If that's referring to me saying 'stop or two better', i thought that RAW was a stop better, JPG was 2-stops better?

Or we'll just have to wait for the 5D3 DxO results...

[sparda79]

I congratulate Nikon on producing a stellar camera in the D800. Now can someone please post sample images taken with both a D800 and a 5DIII, where the D800 made them a better photographer? I have a feeling that I'll be waiting for a long time.

Easy to spot Canon images they are the ones with banding in the shadows & with burnt out highlights  & very poor  reds , I love the Canon red colour its a great red just like to see more shades of red sometimes.

Reds are poor with defaults from ACR in some Canon models, but I found I can tweak settings and get detail without too much trouble.  For more troublesome color images, I use DPP, which usually nails the color.

Totally agree. The ACR defaults handles Canon's red poorly.
Used to be very frustrating...

[Higs42]

So in summary, when you compare the new Nikon that is just hitting the market, it outperforms a Canon 5d2 that was designed in 2008.  Mother of god, who would have guessed?

[psolberg]

So in summary, when you compare the new Nikon that is just hitting the market, it outperforms a Canon 5d2 that was designed in 2008.  Mother of god, who would have guessed?

No. It beats everything recent. And this is the terrible high mp sensor everybody dismissed as being noisy and bad :-)

[V8Beast]

So in summary, when you compare the new Nikon that is just hitting the market, it outperforms a Canon 5d2 that was designed in 2008.  Mother of god, who would have guessed?

No. It beats everything recent. And this is the terrible high mp sensor everybody dismissed as being noisy and bad :-)

Yes, there were lots of Canon fan body in denial, but I think the D800 has proven itself as having a great sensor. To my eyes, however, the real-world benefits of its awesomeness are tough to spot in most of the images taken with it so far.

[Higs42]

The data presented here is for the old Canon cameras.  We will see how the new Nikon compares to the new Canon cameras at some point, but not yet.  So at least keep your comments to reality, not making things up that you hope or wish to be true.  If you have the data for the 5D3 or 1DX, let's see it and we can discuss, but comparing a new camera to 4 year old designs was my point.

[birtembuk]

Numbers are numbers, field is field and I would not buy a gear based on DxO reviews. Just remember the review of the 70-200/2.8II by them. It fares worse than the Sony 70-200/2.8 G...

We are talking here of pure academic assessment by guys in white coat in an immaculate lab. In absolute terms, for sure the D800 sensor will out-resolve the 5D3's. Period. No need whining. Now, just pixel-peep the Still-life 6400 at http://www.imaging-resource.com/IMCOMP/COMPS01.HTM and compare between D800 and 5D3. Not surprisingly, D800 will out-resolve the 5D3 in the fabrics area. This is to be expected as fabrics are sort of micro 3D textures that require lots of resolving power for extracting details. Good if you are in the garment industry. Now, move around anywhere else in the pictures and conclude by yourself. For me, there's a two stops difference in ISO. As far as I am concerned, that's what I am looking for.   

So, what's the big deal here ? ISO or resolution ? Just make you pick and be happy ever after.

[dilbert]

That is for in-camera produced JPEG files only, which means the software in the camera is better, not the hardware.

If that's referring to me saying 'stop or two better', i thought that RAW was a stop better, JPG was 2-stops better?

Or we'll just have to wait for the 5D3 DxO results...

Go and read the Chuck Westfall interview, he is very guarded about what he says with respect to improved sensor performance.

Thus far, all the evidence points to the 5D3 being about the same as the 5D2 for ISO 100 through to 400.

[SRHelicity]

So in summary, when you compare the new Nikon that is just hitting the market, it outperforms a Canon 5d2 that was designed in 2008.  Mother of god, who would have guessed?

No. It beats everything recent. And this is the terrible high mp sensor everybody dismissed as being noisy and bad :-)

Yes, there were lots of Canon fan body in denial, but I think the D800 has proven itself as having a great sensor. To my eyes, however, the real-world benefits of its awesomeness are tough to spot in most of the images taken with it so far.

+1

The data are what they are.  The results are not "biased" (I hear "DxO is biased against Canon" occasionally, which is false) -- they are designed to be objective.  However, what is important is to remember what the numbers represent.  They do not represent ease of use, speed and complexity of the AF system, and in-camera features.  They don't represent the lens ecosystem, nor do they represent the accessories that are available for that particular camera.  Altogether, they DO NOT represent which camera will be best for YOU.  We all have different shooting characteristics (shooting sports vs. portraits vs. landscapes, etc.) and buying criteria (budget, lenses owned, etc.), so a camera with a higher DxO score doesn't mean that it'd be better for YOU or for ME.  DxO results *are* effective at giving us some objective data, however, to allow us to judge sensor performance.

[dr croubie]

Thus far, all the evidence points to the 5D3 being about the same as the 5D2 for ISO 100 through to 400.

And if that's the case, regardless of how well the 5D3 does in high-iso performance, if the iso 100-400 just plateaus again at 12-13 bits DR, then they're pretty much handing landscape photographers on a platter over to Nikon...

[Astro]

And please no, "What if Canon makes a camera like this in a year with more MPs are you going to switch back?"  Because the truth is I wouldn't switch back because Canon will never make a camera like this, not in a long time if ever.

if i had your gift to foresee the future i would not waste my time in a studio with a 35mm.... i would go to vegas!!