Nikon D800E vs Nikon D600 vs Canon 5D Mark III

Over the last weekend, I got a chance to test 4 FF bodies in a field. It was a promotion of the local photo store, and I could have done much better, but unfortunately I missed this chance. Well, at least big part of it.I had something around 1 hour to perform the test, but there were more people around, so I needed to re-set each camera, before I took test shots.

I managed somehow to do it with 3 of those cameras – Nikon D800E (my own), Nikon D600 and Canon 5D MIII.

With Nikon D800 I tried 3 times, but every time there was something I forgot to set, and finally, when I get back home, I realized that D800 was set on a crop mode, I guess 1.2x, as the files from it are 6144 x 4080 px. So even if I do have files from D800, I can’t use them for this comparison 🙁

For some reason I didn’t took the chance to include Nikon D4 and Canon 1Dx, that were both on our disposal. Such a pity, but I was stressed to get all cameras tested within similar lighting condition and to prevent my tripod to move, while I was running to change the cameras to the stand.

So, here are 4 comparison pictures using 100% crops from Canon 5D Mark III, Nikon D800E and Nikon D600, all shot using live view at f8. I used Nikon 24 f1.4 G lens for Nikon and New Canon 24-70 f2.8 L II for Canon. (Another variable, I know…)

First shot is comparing 100% crops from original file size. There is absolutely 0 sharpening (slider in LR4.2 was moved to 0 instead of 25). There is minimal exposure and white balance correction. Files were exported to Photoshop CS 5.1 and crops were aligned in a new document, that I saved in jpeg, using baseline format option and maximum (12) quality.

Canon 5D Mark III + EF24-70 f28 L II top, Nikon D800 E + Nikkor 24 f1.4 AF-S G middle, Nikon D600 + Nikkor 24 f1.4 AF-S G bottom, all at f8, 100% crop – Original file size

Second picture compares same crop but now all files are equalized to the size of Canon 5D MIII (both Nikon files are downsized using bi-cubic sharper – best for reduction algorithm in Photoshop CS 5.1).

Canon 5D Mark III + EF24-70 f28 L II top, Nikon D800 E + Nikkor 24 f1.4 AF-S G middle, Nikon D600 + Nikkor 24 f1.4 AF-S G bottom, all at f8, 100% crop – File size equalized to Canon 5D Mark III – 5760×3840 px. Both Nikon files downsized

Third image compare 100% crops when files are equalized to the size of Nikon D600. (Canon file was enlarged using Bicubic smoother, while Nikon D800E reduced, using Bicubic sharper Photoshop CS 5.1 algorithm)

Canon 5D Mark III + EF24-70 f28 L II top, Nikon D800 E + Nikkor 24 f1.4 AF-S G middle, Nikon D600 + Nikkor 24 f1.4 AF-S G bottom, all at f8, 100% crop – File size equalized to Nikon D600 – 6016×4016 px.(Nikon D800E is downsized, while Canon 5D Mark III enlarged.

Finally, image that compares 100% crops, when all files are enlarged to D800E file size. (Both Nikon D600 and Canon 5D Mark III files were enlarged using Bicubic smoother Photoshop CS 5.1 algorithm).

Canon 5D Mark III + EF24-70 f28 L II top, Nikon D800 E + Nikkor 24 f1.4 AF-S G middle, Nikon D600 + Nikkor 24 f1.4 AF-S G bottom, all at f8, 100% crop – File size equalized to Nikon D600 – 6016×4016 px. (Both, Nikon D600 and Canon 5D Mark III files were enlarged)

And the very last from this test, let’s see, how the files perform when they are enlarged to the D800E, but this time, using my best sharpening procedure. This might be relevant to those who are looking for the best quality in big prints… Of course, the end result might depend much more on the printer and its operator, but in theory, more resolution will give more possibilities.

I used channel + find edges filter, emphasizing edges using curves + gaussian blur + curves again and Unsharp mask set at Amount 200, radius 1.0 and Treshold 0. The difference in retaining details is much more obvious now, and while D800E shows clean detail, other two files start to look smudged at 100%. Of course, there is probably better way to sharpen smaller files than to do so after enlarging them, but the difference will be visible IMO, whatever method will be used, and it is to be expected.

With all flaws, you should take this test with a grain of salt. On the other hand, differences are something we should expect looking at the sensors specification. Will they be smaller or bigger in a controlled LAB environment I don’t know, but the questions is – are they of a significant importance anyway?

I believe (and that was the reason for me to go with D800E, while keeping my Mark 5D MII), that you should consider D800 or D800E only if you really need to print big.

I needed exactly that in a few occasions, and that’s why I choose D800E. (Until I manage to move to MF).

On the other hand, if you are looking for all around camera, and you need higher frame rate for the action, both – Nikon D600 and Canon 5D Mark III would be better solutions IMO, if the resolution would be the only thing to compare within IQ.

However, what seems to me significantly better on Nikon (Sony) sensor, is well known (and documented on respected test sites) – dynamic range. Dynamic range makes a difference in a real life.  Not that Canon dynamic range is terrible (it is very good in fact), but the ability to lift the shadows without noise with Nikon, is simple amazing. (You can often go up to exposure of +3-4EV, or 2EV + lifting shadows at 50%, before any sign of the noise appear).

In respect to Canon, I still like its colors much more, and it usually takes me much less time in PP to bring them to the point of desire. That is however, rather personnel taste.

I would like to express a huge Thanks to the Daniel Drobny from www.fotocz.cz and Karel Gregor from www.fotokurzy.eu, for the great organization of this promotional weekend. If you are living in the Czech Republic, consider to help them staying in business, by ordering something from their shops. Both of those gentlemen, are great professionals and lovely persons.

Here are couple of shots with Nikon D800E from the weekend.