It's not just a question of software-based RAW processing doing "better" than the camera - it's a question of it giving you more control. Every picture is different, every user different. A camera can provide only a handful of one-size-fits-most options.
But I'd say, browse around (see for example Imaging Resource) to find RAW samples from various camera and try processing them yourself using trial versions of the various programs out there. I saw an interesting comparison on dpr where someone took the raw source files images used in the "comparometer" for the K-7, K-x, K20D, and K10D, ran them through ACR (the Adobe RAW processing engine within Photoshop, Lightroom, and Elements) using defaults. I think this was within the thread called "high iso differenceL hype?" or something like that. Anyhow, the results were, if not completely indistinguishable, really surprisingly small. But I repeated the test (well, I used K200d instead of K10D) using ACDSee Pro 3 and found the K-7 more noticeably behind the other three to my eyes. I didn't bother to try to custom tweak the NR settings for the K-7, but what this showed is there is a range of possible results that is software and user dependent.
|