hello Francesco,
But does not that assume the bricks are placed in about as perfect in a raster like the sensor is?
I see a brick wall as a not perfect human shaped raster...
Also the bricks themselves are rough they have broken edges...
Hi Pieter,
There is no need for perfect symmetry to cause aliasing. All that is required is finer detail than the sensor can resolve, and due to the Bayer CFA that resolution limit differs between Green and Red/Blue. If the detail has a high enough contrast (and with a decent lens, in focus, on a sensor without OLPF) it will, that will cause visible false color moiré. It will be even easier to detect if there is a slight difference in angle between the sensor array grid and the subject pattern, because the expected subject pattern is modified/distorted to different directions than the pattern itself.
Anyway, in my experience moiré is a major problem of the ACR/LR raw engine. I have my ways of dealing with it, but it takes time and effort...
Correct, ACR/LR show more False Color moiré and demosaicing artifacts than some other Raw converters. Some of it can be hidden with postprocessing, but one starts with a disadvantage. It also produces slightly lower output resolution than e.g. Capture One or RawTherapee, but most people use LR for the workflow benefits, not for the very best technical image quality.
Cheers,
Bart