- Banned
- #81
Mike Nagel
Banned
It would be interesting to see what ARRI colour scientists and developers could do with the colour filter array channels from RED, for example. It would look different. There's lots of things that are akin to demosaick algorithms that can be done on the image after debayer. For example, denoising is closely related to demosaic.
it would be interesting, but not sure if any useful or better what Graeme provides... would also be interested to see what Graeme would squeeze out of the ALEV sensor... ;-)
debayering and color science is paired and tweaked/adjusted/optimised for a given sensor... the sensor's characteristics are more important than the CS afterwards... the CS can squeeze as much as possible out of it and try to clean up certain problem areas but it can't create what is not there (--> what wasn't captured) and/or it is limited as it cannot know what the real scene looked like... the latter is the real problem if the sensor contaminates the capture and/or smears colors under certain conditions... or maybe the CS is inferior and and distorts the image...
sure, technically speaking u can (try to) do in post whatever u want from the raw but true IQ/PQ and image pop comes from very clean color separation, whatever u do with the image afterwards creatively whether uniform across the frame and/or preferably in isolated areas is a purely creative choice depending on the story the image tells...
the problem is that one does not and cannot know where the image is contaminated (by the sensor and/or CS), whether lightly (uniform) or partially (areas within the frame)... any high end colorist knows that... work off color charts try to get the image closer to the actual scene appearance... u will see a different image now than just working off RLF/DC1-2 and white balance... and u will still now see partial contamination that is there although some parts of the image are aligned just fine and appear very very accurate... but u still may experience greenish and/or unrealistic skin tones...
and surely raw provides almost unlimited potential but I can tell u, u can only bend the image so much... best bet (as always) is to shoot under optimal conditions which mostly means full control over lighting and avoid mixed lighting like the pest...
very interested to see what LI will be able to do as a global tweak to the Dragon capture but since they'll be using a custom OLPF u see already that they're adjusting the sensor capture before the actual A/D and CS kicks in...
I say it'll be good for all of us