this is just a positive feedback for experience with Color Checker.
Recently I had a photo shoot for a prom. I decided to experiment with the Color Checker and calibrate to both sunshine and shade.
While I tend to use a generic calibration for all situations (universal) - I never used it before to a very specific target.
I was surprised how good the results turn out. The main benefit for me was that the colors of the dresses of the kids were quite accurate. This helped me to apply the calibration to all the pictures and mainly focus on brightness, cropping, and some retouches where needed.
I shot roughly 1800 pictures and ended up with about 150 keepers.
Just wanted to say
Thank you to all that work so hard towards creating darktable!
adjusted the exposure based on the color calibration suggestion
Just noticed the profile was even ranked “passable” but to my eyes the pictures turned out very good.
That answers it… It was not clear if you had used the color checker software and made a camera profile ie an icc profile… I guess in theory you could do both the ICC and then improve it using the cc matrix…
There are a couple good videos on doing this…basically you are applying a channel mixer matrix optimized to improve the color achieved using the current input profile.
I have been meaning to try and do first the icc file, ie calibrate the camera and see how much that improves the values as assessed by the CC module and then run the CC module calibration and compare this to doing either alone to see if there is a cumulative benefit or if in fact one or the other is pretty much all you need…
I’ve seen the videos of AP. He points to take picture from the screen first and have the WB adjusted. I’ve tried this in the past but was unable to notice improvements. Maybe there were but not anything that would be a significant to me (based on the experience)
What I did is roughly what the video points to. However - AP points before applying the calibration - to change the illuminant to “as shot in camera”, then apply the calibration and then make a style. To my understanding - the benefit is that we rely on the camera to determine the white balance and the color calibration from the color checker is used to adjust the RGB values. Reasoning was that the day light calibration is good and the swaying is not significant (morning vs noon vs afternoon). I typically use this method and fall back to no calibration when I have evening shots.
This time however - I did not change the illuminant and left it to what color checker would determine because I wanted a calibration that is for this specific event only.
My reasoning was - I was after the skin tones and the color of the dresses and suites.
I’ve had gray (or white) card in the past but my experience was hit and miss. It took me a while to understand that many of these are not that well made and that it is better to use the white because the gray is meant more for film.
In my opinion - having the complete patch card was the best experience so far.
The one time when it didn’t work well - at all was inside shot and mixture of artificial light, flash and Christmas lights but on the other hand - I didn’t position it well either so the light mixture was all over the place.
You can use Argyll or Dcamprof to make an icc from a test shot. Then do as you did…that is the extra step I was wondering if it would help that much…
I did a quick shot with my phone
The embedded profile came up with around delta E of 7 before wb and it was 3.58 before wb when I did a quick icc profile… Then the CC brought that down to 1.5 at output…. So just ripping through and not being careful… it would appear that you might be able to get a boost…it was a lut profile though so I should go back and make a matrix profile… With this icc profile it didn’t calculate the exposure value…. I think it needs to be a matrix profile
As for that picture of the screen, it was really just meant to help get better d65 values to be used by the camera wb… with the new mode of as shot to reference I am not sure that this will be a big deal unless you stick to the previous method and if the values were good to begin with there also wasn’t much to gain…
While the direct sun was with much better delta E - I used it on only few shots (where the shot was under direct sun light).
The 2 shots were taken seconds apart and the only difference was direct sun light vs shade.
My observation was that the sample on the shade produced much more desirable results (on the shade) and the sample on the direct sunlight excelled under direct sunlight.
And this was the main take away for me - to take a separate CC shot as specific to the lighting that I can. For outside shots that would be a minimum of 2 direct light vs indirect.
I briefly searched for dcamproof and it has a lot of praises.
I found other threads on pixls.us and also this page http://rawpedia.rawtherapee.com/How_to_create_DCP_color_profiles
it has a reference to the github repository - there is no packaged format.
I will have to experiment with it (and maybe get over the fear of compiling the software) but most likely - after the Summer. It is time to be outside and taking pictures
