Re: Results: X-AR+LR 6 outresolves X-pro3+Iridient
2
sluggy_warrior wrote:
il_alexk wrote:
You do realize it's a weekend right? Typing this while picking up a dog for a walk with my son from a dog rescue. We are "dog walking volonteers". πππ
Got any dog?
Got a chance to try the X-A5 vs X-T20 in darktable. I've looked at both ISO200 and ISO6400, and agree that the X-A5 resolves details more accurate without artifacts. The X-T20 shot, whether Markesteijn or Frequency Domain Chroma, still have the artifacts at small details.
For example, without any sharpening applied, the text "as a compass" (left box) can be read clearly in the X-A5 shot, but not in the X-T20.

I still prefer the high ISO noise of the X-T20 over the X-A5, chroma noise is slightly less pronounced, but that's just personal preference.
Bayer always does fine text better - kind of An X-Trans bugaboo, but it's not really an issue in real world photography (less of a problem with Enhanced Details). I still think X-Trans does a bit better with other detail (like the green stuff and the etching). Overall though, with sensor resolution increasing, Bayer probably makes more sense at this point - competitive, if not better detail overall, less demosaicing processing required, better LR compatibility, different, but not really much worse moire, and Bayer is probably cheaper to produce too. I have no major issues with X-Trans, I'm used to it, I've got the necessary post processing sorted out and I can produce excellent images with it. That said, I think its usefulness has probably run its course and would very much welcome an X-T4 or 5 with a premium Bayer sensor.