DxOMark vs real world shots?

[jaomul]

It seems a lot of guys with a lot of experience quote the side by sides with results from here. Does anyone think that the science may not always be the definitive guide? A very hot topic at the moment is the 60D/D7000 debate. It's very regularly asked. I am not looking to buy either but intrigued that the D7000 absolutely throttles the 60D,yet I can find many reviews that favour the Canon (also many favour the Nikon). Is it all marketing, can we believe any reviews. I am sceptical when people say "their" brand is better when so few seem to have more than a single loyalty. Curious to what ye think (and I don't care which of the cameras mentioned is better)

 

[KmH]

DXO Mark uses industry standard testing techniques, and is independent.

DXO Mark only tests Raw image quality.

DXO Mark explains all this on their web site.

 

[Derrel]

Does anyone think that the science may not always be the definitive guide?

A lot of Canon users think the science of DxO Mark is bogus, largely because so many of the sensors Canon has developed are being trounced by newer, better-performing sensors found in Nikon, Sony, and Canon d-slr cameras...these Canon honks used to be here on TPF in fairly large numbers but have thankfully moved on to a larger forum that is filled with Canon fanboys who shout down anybody who dares to bring up scientific tests that show their favored products in a less-than-dreamy light.

One of the MOST-critical specifications that DXO Mark can and does test and evaluate is the Dynamic Range that the sensor can handle in RAW mode. That tends to carry across the majority of the ISO range. And that is the area where the newest, Sony-made sensors have made some absolutely HUGE increased in performance over the last couple of years. SOny sells sensors to both Nikon and Pentax, and of course, also uses its own sensors in Sony-branded cameras. So, look at the sensor dynamic range, as well as the color bit depth figures very closely!!! A sensor that has a 2 or 2.2 EV HIGHER (wider) dynamic range is a better imaging tool than one that tests out as being two EV lower...cough,cough..cough...

It is human nature to want to defend one's choices...it often really hurts to see cold, numerical scores and ratings, which show that "your choice" is out-performed by some "other" product. It's easy for people who feel that way to say that science is utterly irrelevant, misleading, or somehow less than "fair".

I currently shoot mostly with a Canon 5D, followed by a Nikon D2x, followed by my Canon 20D as my "knockabout" camera. Two of those cameras have absolute "crap" sensors, but if used correctly by somebody who has good glass [check!] and who shoots in RAW [check!] and who understands how to use the light meter and to expose correctly [check!], pretty good images can be made--within the limitations of the equipment.

 

[dots]

Dxomark provides useful data. Not least the actual ISO ratings for cameras which can explain why a handheld meter is spot-on. I only shoot JPEG and always bracket. No time to piddle around with raw files and Photoshop. Get it right in camera. Delete everything else.

 

[jaomul]

To KmH- thanks for reply, I have seen the site and have an idea how the results are based, I take it from your reply that you would consider results from this site to hold more water than a magazine or website review as it is independent?

Derrell= Thanks for reply. I have to be careful how I write things from here on, I have Canon gear but don't really have loyalty to any brand. Recently I recommended a Nikon D90 to my friend over a Canon 50D as I had used a cousins for a while and thought it produced nicer images than my camera. I enjoy reading about all this technology stuff, cameras and other electronics even if most reviews I read are of things I wont ever buy. I learn a good bit from your posts (very informative), though I suspect your somehow employed by Nikon

Dots- getting it right in camera is a skill I am far from mastering, hopefully in time