6 replies to this topic

[loftus]

For pixel peepers, an interesting review of the DxO score process

http://www.luminous-...g_cameras.shtml

[Paul Kay]

Well I've just read it and I still have no idea whether you can actually SEE the differences between the better cameras in a final image.

[loftus]

I think we can see differences as cameras get better, maybe not in small increments; but certainly over time - for example cant even compare the RAW quality of my D200 with my D7000. Having said that I think it is at the margins - highlights and shadows, high ISO etc that the differences are more evident, and not so evident (or not evident at all) in the midtones, normal ISO's etc.

Edited by loftus, 17 February 2011 - 04:24 AM.

[Paul Kay]

I agree that sometimes we can see subtle differences between one camera and ins successor and over time there is no doubt that technical 'quality' has improved. But these pseudo-technical 'tests' and scores seem to present highly complex data in simplistic terms and simply bringing it down to a few graphs and numbers might be appealing but actually tells us very little in terms of the effect on the actual images produced by the tested cameras. Personally I think that we are getting to the point of lens limited images and the cost of producing lenses capable of showing significant pictorial improvement is going to be high (I've discussed this with a lens designer friend several times). The problem is that many people buy the perceived 'best' based on reviews, tests and scores instead of actually looking at their requirements and checking to see if equipment is capable of meeting them. I usually suggest that such tests and scores are ignored if I'm asked about cameras and someone quotes them (asking "what do you think they mean?" may be a little unfair but reinforces the point).

[loftus]

I agree with you; if anything these tests reinforce how good ALL the cameras are, as one can see that frequently the scores are pretty close. If anything they dispel the myth that one brand is significantly better than the other in picture quality.
But then I did say that this was for pixel peepers....

Edited by loftus, 17 February 2011 - 05:45 AM.

[Paul Kay]

I agree with you; if anything these tests reinforce how good ALL the cameras are, as one can see that frequently the scores are pretty close. If anything they dispel the myth that one brand is significantly better than the other in picture quality.
But then I did say that this was for pixel peepers....

Oddly enough I'm going through loads of files at the moment (hence office and computer bound) and I keep thinking that 5DMkI files are really nice - the 5DMkII files don't look quite as good - and then I realise that I'm not comparing like with like and viewing anything at 100% is not going to make it look particularly good. Both cameras produce excellent images and if I'm honest the MkII is overkill - but was bought for its video capability and has more than paid for itself. I do like the original 5D though. And whilst on the subject, I'm still using files fro the 1DS and S2Pro........ they all reproduce pretty well at the sizes most demanded from them in publications.

[bvanant]

Well I've just read it and I still have no idea whether you can actually SEE the differences between the better cameras in a final image.

My guess is that a lot will depend on how you are looking to see the differences. If you are looking at a 1000 or 1500 pixel wide photo on the web, I don't think you can see the difference. On a very large monitor though, you can begin to see the differences. I can see (or I think I can) differences between my 50D and 7D at large enlargements (shot with the same lenses). In the lab we can definitely see differences between the 5D and 5DII when they are attached to a microscope so the differences are there. If you make big prints I think you can easily see how much better the big prints from newer cameras/larger sensors are than from older/small sensors (with the same lenses). For much of what we do though, particularly for looking at stuff on the web or on a typical 20 inch monitor it might be tough to show.

Bill