> > You didn't question any claim, you posted a snotty little obtuse comment > that read so: "Really? What do you think does the conversion to JPEG, > then?" Therein lay the principal instance of nitpicking in this thread. > You're just miffed because you're used to getting away without being called > on it. > > Seeing that you had been caught playing the pedant, you then switched your > position, acknowledging that there could be a difference, but that you > doubted that the second method was *better*. > > You then MANUFACTURED a quote to make your own, newly adopted, position seem > more reasonable, and go on to take me to task for noticing. HAR!
Nope. I haven't changed my position on this, so any mistaken belief that I did so as a reaction to your 'calling me' on it only exists in your mind. The firmware is software. It comes from Pentax - the same supplier (to us) as the software used in the Pentax Photo Laboratory. If there are complaints about one implementation, then the second implementation at least bears examination, possibly even more so since that implementation has to live with the additional constraints of working within the restricted in-camera embedded system. That's where I started out. That's where I am now. If you don't like the way I originally asked the question, fine. Now talk to the issue. The software implementation appears to use different JPEG parameters, as the resulting file sizes *seem* to be different. (hard to tell; there's no way for us to run both the firmware and software algorithms on the same input). But JPEG conversion is only part of the story (and probably not the most important part). There's also the interpolation from the raw Bayer matrix values, and applying any image processing filters. These may be using the same code in both implementations (complete with any bugs), or they may be total rewrites. We've seen claims that third-party packages can do considerably better than Pentax Photo Laboratory when processing raw images. We've also seen people say that in-camera JPEG seems to be comparable in quality to JPEGs produced from raw files by Photo Laboratory. Does that mean that the in-camera JPEGs have the same problems people see from the software-converted files, or does it mean that different standards are being applied by the different testers? I don't *wan't* to find out that in-camera JPEG conversion suffers from the problems Photo Laboratory apparently has, but if that really is the case then I want to know about it so I can decide whether to use raw mode more often.
