Originally posted by RiceHigh I don't read Japanese nor do I trust automatically generated translations, but looking at the pictures, I might assume the text is saying, "see here, the K20D is the only camera that didn't blow out the highlights". Which is to say, it exposed perfectly, the others overexposed. And not just a little - they overexposed *horribly*.
It's obvious enough to the naked eye, but if you really need proof, download the E-520 sample and look at the histogram. Can't you see that giant spike along the right side showing how badly the image is clipped? If you still have any doubt, click an eyedropper anywhere in the broad white building (at least, it looks pure white in the E-520 picture, because it has overexposed so badly) with the word "Conrad" on it. I don't think there is a single usable pixel anywhere on the face of that building - they are virtually *all* maxed out at 255,255,255. The E-520 managed to complete wipe out all detail on that building - it looks like a solid mass of white. The others were only marginally better, showing only the faintest hint of the brickwork. These are almost *comically* bad.
The Pentax picture was the *only* one not to blow out most of the detail! And if you look at the histogram, you;ll see it it is within a quarter stop of clipping. meaning it completely *nailed* the exposure here - it could not have exposed any more without ruining the picture a badly as the others did.
I mean, really, if you wanted more definitely proof of how *great* the Pentax metering is, don't think you could do better than this example! Please, for anyone still harboring even the slightest doubt, clicking on that link and then view the K20D picture from the first set.
The fact that you can look at those pictures and not see just how badly the other cameras are overexposing the scene and that Pentax has nailed it perfectly is but one piece of evidence that you do not understand the first thing about exposure.
Quote: Also, DPR technical writers are not knowing well how to use the K20D as they applied the most exposure compensations for the shots they made, when they tested the K20D!
And the fact you believe that needing to apply exposure compensation for some scenes is indicative of a problem is yet more evidence that you don't understand the first thing about exposure.
But thank you for posting this link to the perhaps the most conclusive proof I've ever seen of just how bad other camera's metering can be, and just how off-base people can be in claiming Pentax is underexposing.