D3 vs mk3: A Single Fast Comparison
Today, while shooting OSCON 2008, I got the chance to handle a Nikon D3 for a few minutes, courtesy of Ted Leung who has his copy with him this week at the conference. Seeing a quick opportunity to shoot a comparison shot, I put my compact flash card into the Nikon and matched a shot I made with my Canon 1D Mark III with the same exposure settings at ISO 1600.
The shot is of a whiteboard in one of the conference rooms under what can uncategorically be called crappy light. The above shot is from the Nikon. There’s not much point at posting a 500 pixel wide image from the Canon here. Instead, here are crops of the resulting images at 100%:
A few notes about this crop: both images were processed with Adobe Camera RAW via Lightroom. White balance was adjusted to match, but no further processing was performed. No sharpening. No nada. The crops are saved as a 24-bit PNG file to avoid compression artifacts. And, that extra blue bit in the right side of the Nikon image is another letter on the whiteboard in the background that was exposed by a slight angle change between shots.
This is most definitely not a rigorous and exhaustive comparison. In fact, it’s so quick as to be laughable. Therefore, instead of saying much about the crops, I’ll let them speak for themselves and only add that it pretty much matches noise comparison results I’ve seen elsewhere. Maybe at some point I’ll have the chance to really compare the systems more in depth to each other, but until a full Nikon kit lands on my doorstep or I can steal a lot more time with Ted’s camera, this will have to do.
Related Posts:
5 Comments
There is absolutely more noise in the MkIII, but there appears to also be some aliasing likely due to the 2MP difference in the two systems.
Reply to this comment
Obvious comment: it doesn't cater for lens variation. Granted, that's part of the appeal of the Nikon system at the moment - the 14-24mm f/2.8 would be a sweet, sweet lens to put on the 21 MP 1Ds III - but nonetheless. If you want to compare noise to noise, you should really be putting the same optics on each camera (possibly a decent Sigma lens?)
Still, it's an impressive sight. It will be interesting to see what Canon comes out with in the next year or so.
Reply to this comment
Did the Canon have the same aperture and focus point, with the same equivalent lens? I know it's just a quick comparison – I'm not overly poking holes in the method, honest! – it just seems like there's more to it than the extra noise: the 'ship' of membership looks distinctly softer in the 1D Mk.3, with more fringing. Perhaps it's a difference in resolution, or quality of lens, or even a bit of hand-holding variation?
Anyway, all that aside, the D3's results look truly incredible from what I've seen so far, the (lack of) high-ISO noise really is remarkable and puts Canon to shame. I'm sure there are many Nikon system users that are overjoyed at this, as Canon have consistently stolen the march for several years in the digital race. I'm intrigued as to how they'll respond and would've though that the upcoming Olympics would be the perfect venue to showcase some new prototypes, perhaps along with the 5D successor. We'll see (I hope)!
Reply to this comment
The aperture, shutter, and ISO settings were the same on both cameras. Aperture was f/4.0. White balance was different, but equalized in Lightroom. Both cameras were sporting respective 70-200mm f/2.8 lenses with IS and VR enabled. I attempted to hit the same focus point, and it appears I succeeded, but variation surely exists in some fashion.
As far as the sharpness differences go, that could be lens issues, hand holding issues, luck, etc. I wouldn’t read too much on this front based on this quick comparison. Too many unconstrained variables. For example, Ted’s copy of his 70-200 might be especially sharp for Nikon and the 70-200 I’m using might be a bit soft. I don’t particularly think this is the case, but it is a possibility. Probably the only fair thing to judge here is noise as that should be lens, focal point, and stability agnostic.
If I had access to multiple Canon and Nikon bodies and lenses to compare, then we could really start having some fun. Maybe Nikon can loan a bit of kit my way? And, of course, maybe our anonymous Canon employee friends can arrange to drop some next-gen Canon kit off as well so that I can have comparisons ready to go when they announce. And, maybe pigs can fly... :)
More seriously, maybe something more rigorous involving tripods and what not can be arranged later in the week. Unfortunately, I am working the show this week, so might not really have the time.
Reply to this comment
Actually, to really highlight the awesomeness of the Nikon sensor you should have used something more sensitive than 1600. I got a D700 yesterday (smaller camera, same sensor) and the low light performance is just rub-your-eyes unbelievable.
- ask
Reply to this comment
Leave a comment