D3 vs. D2X test shots

[shoyles]

I know, these sort of tests are all over the web but I had to see for myself how the D3 stacks up against the D2X.

The following shots were taken in manual mode F/5.6, 1/250s, off camera SB800 bounced off of the ceiling (also in manual). I used a 70-200mm F/2.8 VR and adjusted the focal length to account for the difference in sensor size (stood in the same spot and zoomed to the same composition).

The cameras were hand held and focused (auto) on the memory card in the middle of the following test subject.



Here are the 100% crops at various iso settings. These were all done in RAW and converted with lightroom at default settings with equal white balance. No extra sharpening or NR applied. EXIF data is intact.

ISO 200 D3 / D2x



ISO 800 D3 / D2x



ISO 1600 D3 /D2x



ISO 3200 D3 / D2x



ISO 6400 D3 / ISO 800 D2X



To me, the color on the D2X looks warmer and the exposures don't match (d3 is hotter). The D3 also has a narrower depth of field (under these conditions). There is definitely a difference in the way it handles high ISO. Looks like about a 3 stop improvement to my eye.

[jeffkohn]

Hmm, at lower ISO's the results are somewhat mixed here. At ISO-200 shot I prefer the D2x shot, as it has more texture. At 800 the D2x is starting to look a bit grainy, but still pretty film-like. The first two D3 shots almost look too smooth (dare I say Canon-esque).

Of course the D2x is terrible at 1600 and up, there's no comparison there. But in these examples I don't think the D3 6400 looks as good as the D2x 800; the D2x noise looks more like film-grain while the D3 is cleaner overall but a bit splotchy looking in the shadows and shows some chroma noise. Still, very impressive for ISO-6400.

I have to wonder what the default NR settings are in LR, because your ISO-800 D2x shot looks cleaner than anything I ever got with my D2x at that ISO. And the D3 ISO-6400 shot looks to me like it's had some NR.

[shoyles]

Originally Posted by jeffkohn

Hmm, at lower ISO's the results are somewhat mixed here. At ISO-200 shot I prefer the D2x shot, as it has more texture. At 800 the D2x is starting to look a bit grainy, but still pretty film-like. The first two D3 shots almost look too smooth (dare I say Canon-esque).

I think the D3 shots are simply all a little over exposed compared to the D2x. This gives the D2X more texture. I'm not sure what is causing this as I ran everything in manual and the same camera to camera.

Originally Posted by jeffkohn

I don't think the D3 6400 looks as good as the D2x 800; the D2x noise looks more like film-grain while the D3 is cleaner overall but a bit splotchy looking in the shadows and shows some chroma noise. Still, very impressive for ISO-6400.

It's hard to compare shadows of ISO 6400 with ISO 800 as the room lights were messing with me on the ISO 6400 shot. I had to turn the flash down to 1/128 power and move it to the other side of the room. This made the shadows different at the high ISO settings.

Originally Posted by jeffkohn

I have to wonder what the default NR settings are in LR, because your ISO-800 D2x shot looks cleaner than anything I ever got with my D2x at that ISO. And the D3 ISO-6400 shot looks to me like it's had some NR.

Lightroom noise settings for all shots...
Lum = 0
Color = 25

[AndrewCCM]

DOF is different too.. Notice how much sharper the text is on the right (D2X). I am sure that has to do with the zoom and FF sensor, but it does affect the perceived sharpness in the test.

Thanks for posting..I am always interested in independent testing...

[Riff Raff]

Wow, that's terrible on the D2X. I'd always heard that the high ISO was bad there, just never actually seen it. It certainly never sunk in that my low-end Digital Rebel was competitive with the Nikon pro cameras in that respect. Pre-D3, of course. Better competition is always good for the customer, in the end.

[shoyles]

Originally Posted by Riff Raff

Wow, that's terrible on the D2X. I'd always heard that the high ISO was bad there, just never actually seen it. It certainly never sunk in that my low-end Digital Rebel was competitive with the Nikon pro cameras in that respect.

Agreed, Canon has always had better high lSO performance than Nikon. I have done similar tests with the 30D and it beats the D2x by about 1 stop. The D5 is a little better than that (ca 1.5-2 stops). The D2X is still a great camera. It has outstanding image quality and autofocus. In fact it has the best low light autofocus I have seen on any camera (with the exception of the D3 / D300). You simply have to keep it below ISO 800 if you want to make big prints. With the release of the D300/D3 Nikon has changed all of that. Hopefully Canon will up the ante and give the market something that blows the D3/D300 out of the water when they release the 5D MKII. This will make Nikon counter with a better camera and so on. Gotta love competition.

[Metalhead]

well, as I see it the DOF isn't going to be the same, anyway. For the test to be 100% accurate, I think, you'd have to have the focal length the same on both lenses, and simply move the camera to get the desired composition. The DOF will change at different focal lengths, and the D3 will probably have more softness anyway, since the sensor reads so much farther out than the D2x. That's just my opinion, though

[PictureCow]

Thanks for posting these Steve!!!

[shoyles]

Originally Posted by Metalhead

well, as I see it the DOF isn't going to be the same, anyway. For the test to be 100% accurate, I think, you'd have to have the focal length the same on both lenses, and simply move the camera to get the desired composition. The DOF will change at different focal lengths, and the D3 will probably have more softness anyway, since the sensor reads so much farther out than the D2x. That's just my opinion, though

Good point. There is no perfect way to compare cameras with different sensor size. If I moved in and zoomed out I would have also narrowed the depth of field because I would be closer to the subject. I don't know which narrows the depth of field more in this case, focal length change (my test) or distance from subject. Any thoughts?

[Metalhead]

Originally Posted by shoyles

Good point. There is no perfect way to compare cameras with different sensor size. If I moved in and zoomed out I would have also narrowed the depth of field because I would be closer to the subject. I don't know which narrows the depth of field more in this case, focal length change (my test) or distance from subject. Any thoughts?

I'm not sure, but I think the focal length would make a bigger difference

[jeffkohn]

For a given field of view, the larger sensor will always have less DOF. Moving closer versus changing focal length will offset each other. If you want the same DOF at the same field of view, you would have to stop down the FF camera to a smaller f-stop.

[shoyles]

Originally Posted by jeffkohn

For a given field of view, the larger sensor will always have less DOF. Moving closer versus changing focal length will offset each other. If you want the same DOF at the same field of view, you would have to stop down the FF camera to a smaller f-stop.

or keep the focal length, aperture and distance the same and just crop. Of course this will kill the resolution. I guess there is just no winning.

[mramos]

Congrats Steve on the new camera. Sorry for all the happy faces, but this is how I feel since I might actually get a chance in the near future to shoot with it.