DxOMark launch raw file image quality database

http://www.dxomark.com/

Rather strange marking when comparing the K10D and K20D

Check the K20 and GX20!

Hmmmmmm .....

GX20 15mp and Iso 3200?

Interesting comparison between Pentax K20D and Nikon D300.

First one I did was K10D Vs D300 Vs 40D

If you compare the K20D to the Nikon D300 and the Olympus E-3, the Pentax has nothing to be ashamed of.

does anyone else find it interesting that the K10D seems to beat the K20D in almost every class on that test?

I thought the K20D was widely regarded to have better noise performance and dynamic range than its predecessor?

Have I interpreted correctly that K10 and K20 are essentially the same apart from the fact that the K10 actually has considerably better dynamic range?

that's certainly what it looks like on those charts, but my point was that I don't believe that's what people have been reporting in real life.

of course it could be people who have spent large amounts of money not wanting to believe they made a booboo, but surely at least one person would have turned around and said "hold on, this is no better than my K10D and the DR is actually worse".

Anyone own a K10 AND a K20? fancy sheding some light?

Of course there no obvious test of resolution in there.......

.....Resolution had BETTER be superior, otherwise all we have is a facelifted model......at best!

Quite a lot of DSLR buyers don't really have a clue about ANY of the technicalities of image quality and are interested even less, especially if it means they basically might have bought a pup!

The test results are not really that conclusive in that most are more or less identical, certainly within the limits of experimental error.

The dynamic range does seem to be different, but then the resolution advatage probably more that outweighs that.

It won't stop us buying a couple of K20D cameras very soon. Chris can expect a call....

Yeah, I'm puzzled too. I've worked with: K10D, K20D, Canon 20D and 30D RAW files and find the following (in real life):
K10D produces better results than either Canon
K20D produces better results than all three... and especially at higher ISO. The DxO results look really odd! DPReview cited at least 1 extra stop of dynamic range in the K20D over the K10D... and that seems to be what I've found.

Checking the DxO results it looks like I should ditch my gear and go for a Canon 5D or Nikon D700

Read into it what you will
Matt

Perhaps they got the results mixed up. Wouldn't be the first time, and reviewers are paid peanuts. To make the job worthwhile financially, they have to move quickly and cut corners.

G

I wondered the same thing......

While I was driving just now I was turning it over in my mind and wondered if they have got the graph plots or colours round the wrong way.

I don't think they can have, since the K10D can't reach ISO 3200 and 6400