Originally posted by MMVIII Exactly. I based my opinion on this topic exactly on such a comparison. It is obvious that the theoretical, even if measurable by fourier transformation analysis, "loss of detail" is something that would be almost impossible to observe in everyday photography and is masked by multiple parameters by a high factor. The focusing, resolving power of the lens and by a huge amount by the demosaicing algorithm of the raw convertor. The last does produce horrible results for the Sony, that should have an resolving advantage due to the sensors resolution, which is in some parts completely lost.
This seems to be also the biggest source of disagreement between the reviewers and the actual users. The reviewers jumped to conclusions after they thought they nailed the "detail loss" by using Bill Claffs analysis and supporting it with images. When it became obvious that the images were technically insufficient and could not be used to support such a claim they started to get defensive and still insisted that this sensor would be a "star eater on steroids", which also turned out to simply not being true. That it would be practically impossible to see "data loss" is supported by the fact of the initial assessment of the KP image quality and the praise for it "boxing in a higher class"...
Now there is only the fact, that some are still following the idea of "pureness" of the raw, and they see this endangered by the K1IIs implementation. But this is a matter of principle and probably not possible to come to any common conclusions on that.
Trying to sum it up:
If your photos would suffer or become only slightly less usable because of the accelerator unit you should be alarmed and have changed multiple other parameters before you could blame this chip.
^One of my favourite posts, thanks for posting.
One of my first big endorsements for digital photography was an aerial photographer who shot images of people's cottage lots for their walls. The fact that he'd switched to digital for the resolution, which in his case was all important, let me know, sooner or later I was going to want to.
What many don't understand is that, people who work with the technology actually have a much better understanding of what these things mean than website camera reviewers who essentially aren't qualified to babysit your kids, forget about understand the parameters necessary for camera evaluation, and the limitations of such evaluations.
The only thing that would make these guys more believable would be a thorough description of the limitations of their evaluations like the side effects for medications. Essentially, if you don't know where you're going wrong, you have no idea how to improve your evaluations. Web site analysts are little better than snake oil salesmen.
What's more irritating than that is people who go looking for a problem, who do something no photographer would ever do to demonstrate a "problem" then claim there was something wrong with that the manufacturer did. If you discover a problem, that no one will ever come across in daily use, honestly, no one cares.