Tested D3 vs 1Ds mk2 today

Started Nov 4, 2007 | Discussions thread
Flat view
Benedict Slotte Senior Member • Posts: 1,165
Tested D3 vs 1Ds mk2 today

I visited the Digiexpo fair in Helsinki today, and had a chance to try out the D3 and compare it to my own 1Ds mk2 that I had with me. I would have liked to add the 1D mk3 as a third camera, but Canon didn't let people insert their CF cards into the cameras (not even the 1D mk3 although it's already on the market...!). What a shame - because at the Nikon stand people were actually encouraged to insert their own cards into the D3 and shoot away. Also the 1Ds mk3 would have been a nice addition, but I had already expected that I wouldn't be allowed to insert my card into that one.

I know the 1Ds and the D3 cameras are intended for somewhat different purposes and shouldn't be directly compared - but I own the 1Ds mk2 and therefore it was a desired point of reference for me.

I specifically wanted to test high-ISO noise, nothing else. Before you draw any conclusions from these images, let me first say:

  • Do NOT judge sharpness or focusing or white balance or overall colour balance from these samples. They weren't controlled in the manner required for such tests.

  • Judge ONLY noise. The noise seen in these samples is the noise in the in-camera JPGs (depending on sharpness setting too, of course, but I used 0 for the Canon images, and I assume the D3 didn't have any excessive sharpening either).

Here is an overview showing what I shot (a part of the ceiling and installations, basically):

In the following images, the upper row is the D3, and the lower row is the 1Ds mk2. The columns represent ISO 800 (left), ISO 1600 (center), and ISO 3200 (right). Each image is a 100% crop of a given detail.

In lighter tones, the difference in noise between the cameras is surprisingly small, I would say. Also, the D3 shows about as much low-frequency noise (i.e. those large speckles, not the fine grain) as the 1Ds mk2:

Darker tones seem to show about a 1-stop advantage for the D3 (maybe the D3 specifically applies more aggressive noise reduction to DARKER tones in the in-camera JPG). To put this 1-stop difference into perspective, it's perhaps worth pointing out that roughly 1 stop is what one could expect simply from the larger pixel pitch of the D3.

These ones show both light and dark tones:

This image shows an artifact that plagues the 1Ds mk2 (and, I guess, most of Canon's somewhat older cameras): strong highlights leave behind some dark bands. I didn't see this in the 5D (at least not nearly as much), and I would guess it doesn't happen in the new 1D cameras either. In this particular case, the D3 wins completely over the 1Ds mk2:

Nikon also had a few printed samples there, and while their shadows were wonderfully smooth, I saw surprisingly much low-frequency noise in the lighter tones in some of them.

This was a quick and simple test carried out in a short time in between many customers who wanted to try cameras and lenses. By the way, I don't remember which lens was used on the D3. My 1Ds mk2 used the 24-105 mm IS. On both cameras I exposed by spot metering on the white wall close to the center of the image. Now, don't start complaining about how I should have equalized the white balances and ensured optimum focusing etc. etc. because that was not the point of this quick test. And I'm an engineer, so I do know how scientifically accurate tests have to be made

Now if I could compare the RAWs (I saved them too). Is D3 RAW supported by any converter out there yet (I suppose not since the camera isn't available for some months)?

-- hide signature --

B. Slotte
Turku, Finland
See profile for equipment

Flat view
Post (hide subjects) Posted by
(unknown member)
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow