Next play raw. I thought about letting my own edit aside to not influence you. But in the end on the last play raw there were enough creative edits anyway. So I decided against.
IMG_5392.CR2 (26,9 MB)
This file is licensed Creative Commons, By-Attribution, Share-Alike.
A second try … when looking how to “dodge and burn” in RawTherapee I found retinex as one possibility. It give some (too much?) punch to the result.
So here it is
Added +1EV when converting the raw with Filmulator. In GIMP I used luminosity masks to make contrast adjustments and slight changes to saturation within the masked layers.
Very difficult to choose the composition I prefer, so here are two edits. I like the greens and the haze in the background.
IMG_5392.jpg.out.pp3 (14.6 KB)
Not that much going on in the left side of the picture. I was surprised how different it looks when cheating for educational purposes, balancing the deserted barn as an entry point vs. the village.
Done with ART and a module I’ve been playing around with.
I’d like to question the RawDigger graphic posted in #12. Taken at face value, it appears to suggest that this photo is underexposed by about 3.5 stops. I don’t think it is even possible to get 16 bits of resolution from 14 bits of linear data. The camera’s specular white level seems to be about 12000, so if the exposure were to be increased by 3.5 stops, it would result in a blown out mess.
This one, #12:
(Post #5 has apparently been deleted. It seems that if Javascript is enabled, the deleted post is not counted in the fancy sliding bar thingie on the right. Without Javascript, the original numbers are shown in the post, just to the left of the date.)
I think adding some context and explanatory text would benefit the viewer, as would removal of the extraneous and misleading parts. The explanatory text would be helpful to those of us trying to understand your point.
For example, the items I mentioned previously:
If the EV0 marker is the reference value for the maximum data value, why are two stops of (unattainable) range shown “to the right” of it? (The implication is that those 2 stops could have been used for more exposure.)
Why was the EV0 marker placed where it is instead of in a location that corresponds to the range of values the camera actually uses? (This again implies exposure range that is not available.)
So, from my perspective, if the point isn’t that the photo is underexposed by 3.5 EV, some descriptive text about the graphic would help communicate what that point is.
The way I interpret the raw data histogram, it can be argued that the photo is underexposed by less than 0.2 EV. If, as armchair critics, we want to push the photographer’s luck, there’s a maximum of 0.8 EV available if we care about color shifts, or 1 EV if we don’t care about color shifts. Is that really worth worrying about?
(I apologize for derailing this thread, but I do not believe that the photo is underexposed by any significant amount.)
Is their even such a thing as “underexposure” in a raw file? There is clipping and there is clearing the noise floor, and beyond that it doesn’t matter so much.
