Same methodology as before, I used the D800e as my baseline and model for comparison.. I'd have used a 645D but, IMHO, the difference between the current 645D and the D800, the 645D produces a better image, but you have to squint to see it. The 645D really needs to go to 70MP, to be worth more than the D800e.
SO I cropped an Image Resources 100 ISO image from a D800e, then did the same crop on a Canon 1D x, A Nikon D600, and a K-5 IIs. Then I expanded the other smaller MP count images to be the same size as the D800e image. The D800 is so much the winner, it really is in a class all it's own, predictable and unsurprising. What is surprising is the other 3 are very close, and you could make a case that the K5IIs is the best image of the other 3. So comparing a $7,000 camera and a $2,000, the $1,300 camara gives you arguably the best image, at least in these test conditions on this one small portion of a frame. Still the best apart from the D800. Also confirming my suspicion, that as an upgrade path, it's D800 or nothing. Even if you argue that the others are marginally better in IQ, it's not enough difference to be worth money.
The surprise for me, is the D600. I had hopes for that camera. Theoretically a 16MP APS-c shouldn't be in the same ball park as a 24 MP FF. Both images are enlarged not reduced, so the K-5 is actually expanded by a larger amount and by theories expressed many times on this board, should suffer as an image, from being enlarged more. I'm not seeing it. The theory says one thing, the images say another.
Last edited by normhead; 11-03-2012 at 07:40 AM.