Dpreview studio test shots (including ACR conversions) now published

Started May 7, 2012 | Discussions thread
tdptdp Regular Member • Posts: 496
Re: Noise Reduction in "RAW" samples

Andy Westlake wrote:

Option 2 is probably closest.

The 'problem' here is simple: the X-Pro1's completely different CFA demands an entirely different demosaicing algorithm compared to conventional Bayer sensors. So direct comparisons are always going to be tricky - you're not quite comparing like with like. It's also entirely possible that Adobe has borrowed the demosaicing algorithm from Fujifilm - producing another, entirely new one would make little sense given everything else they also have to cover (most notably DR modes and lens corrections).

As it happens, SikyPix conversions of these RAWs look very similar to those from ACR - there's remarkably low chroma noise even when all NR sliders are set to zero. (BTW you can confirm that chroma NR was set to 0 in our conversions easily enough, by looking at the file info in Photoshop - 'Advanced' tab, 'Camera Raw Properties'). Basically, the conclusion has to be that RAW conversions from the X-Pro1 just look different to conventional Bayer cameras in ACR, with lower chroma noise, similar to engaging a degree of chroma NR normally.

Overall, it is what it is.

Fantastic work, Andy.

If you get the urge to answer a couple of slightly off-topic questions, why the 85/1.8G for the D700 and D800, and why no MLU?

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow