You are comparing your programs not RAWs
3
Marco Cinnirella wrote:
Looking through raw images in both LR and C1, I have to say that while I like using the newer cameras due to improved functionality, such as AF, my subjective assessment is
You are comparing the differences between your two pieces of software - namely:
- different input profiles for different camera sensor generations,
- different demosaicing algorithms and/or maybe different software implementations of one single algorithm,
- different default tone curves applied,
- maybe some other implicit default corrections (exposure, WB etc.)
RAWs themselves are color-agnostic and do not contain what we call "image data". RAW is not an image and does't contain any (circa preview JPG) until developed.
So you are comparing your programs not RAWs.
If you want to compare RAWs, you need to
1. prepare a set of input ICC profiles for each RAW breed, with profiles carefully calibrated to identical results using the IT8.7 color target (you can order a good and affordable one HERE, you need the "CF" target ) and appropriate profile-cooking software - darktable-chart free utility from the darktable 4.0 disribution comes to mind first,
2. use the software that doesn't apply any hidden default implicit corrections to the image which are not under your control and you may have no clue about their presence - this it immediately leaves out all proprietary software like any Adobe et. al. because you never know what it does under it's locked hood today; use some transparent open software, like darktable 4.0 or RawTherapie.
3. use the same demosaicing algorithm at same settings for all your RAWs - the best of the breed is the Markesteijn algorithm at maximum passes + VNG
4. apply identical exposure corrections and WB corrections
5. don't apply any sharpening and/or denoising
6. apply identical tone curve to all RAWs.
Actually the whole experiment described is meaningless in practice, because you will get identical results from all your RAWs with the only minor difference being in noise level and fine-grain resolution - sensor with less Mpx is less noisy but provides lower Nyquist frequency.
-- hide signature --
All I post is my own, humble, personal, subjective and highly biased opinion. It may change in time upon new facts and convincing arguments arrival.