Originally posted by Kunzite Any circumstance where your images are made noticeably worse - and not improved - by this accelerator are extraordinary.
Not when taking the superficial view that "starting with ISO 640, there will be loss of detail in raw files". There is nothing extraordinary about "ISO 640 and above". We all know that DPReview exaggerated enormously and that they are not using the same measuring stick for all brands, but technically their statement is not incorrect.
Originally posted by Kunzite Pentax' only fault was to release a camera in two versions, with and without the accelerator.
It took someone outside DPReview's staff to recognise the raw manipulation in an objective manner but the respective demonstration does not require a "before" (K-1) and "after" (K-1 II) comparison. You can tell by looking at the K-1 II in isolation that image processing is being performed.
Arguably, you could also say that a further "faults" according to DPReview were to release the K-1 II with so few differences to the K-1 and that one of the differences was step back (mandatory noise reduction), the other was pointless (hand-held PixelShift), and the third one rather minor (better AF-C performance). DPReview actually used the fact that a K-1 could be upgraded to a K-1 II as an argument against the K-1 II in the sense that it shouldn't be so easy to turn an older model into a new one.
I'm not defending them, far from it! All I'm trying to illustrate is the broad attack surface Ricoh provided. We know that DPReview will find a way to wax lyrical about their Sonys and Fujis and explain with loaded language how DSLRs are not contemporary anymore, no matter what Ricoh does. However, I still think it makes a difference of how easy one makes it for them.