To introduce the topic, I decided to demonstrate how I typically use filmc in my everyday use of darktable.
For me, filmic is always embedded in the processing steps and serves mainly as an aid in dealing with different dynamic ranges of photos one is confronted with as a photographer.
Accordingly, I roughly divide photos into three different groups:
Low dynamic range photos - low contrast, “flat” looking photos with not much difference between shadows and highlights. Filmic can here extend the range and increase contrast.
Photos with “optimal” dynamic range - Well-lit, high-contrast photos that don’t need much processing. For this the default setting of filmic is sufficient and I don’t have to adjust it during processing.
Photos with very high dynamic range, which are mostly underexposed during shooting to avoid overexposure. Overexposed photos (if they are not of particular importance) end up in my trash.
High dynamic range photos can be very challenging to process because you don’t have a lot of room to maneuver with exposure compensation and you always have to watch out for the highlights. Filmic is very useful here for compressing the dynamic range, giving you much more flexibility to emphasize the main subject without risking image artifacts.
So, my approach is that the brightness of the photo and the dynamic range is always subordinate to the subject. This means that I don’t have to use the full dynamic range if it’s not necessary.
And what the subject is, I decide as a photographer and not some algorithm that mathematically balances the dynamic range evenly!
So now for each group a pair of examples. Each time before I start processing, I will move the mouse to the subject I want to emphasize. I will also point to the histogram. I will not only focus on the filmic, but embed it in the processing so that you can compare the final result with the unprocessed photo.
Low dynamic range photos:
Optimal dynamic range photos:
High dynamic range photos:
