I have an issue when editing a larger collection of photos, for example, from holidays. I edit photos one by one, and of course, at the end, I end up with a mix of different ‘styles.’ What is your way to solve this issue? How do you edit photos so that they are consistent together? I’m adding an example of some photos from last year’s vacation.
I had the same issue until I bought a Fuji camera and used out of camera jpegs. It is really the only way I was able to achieve consistency.
At least try to limit oneself to one camera is helpful.
I would love to hear from others how it is possible to get one “style” from several different camera in various different situations.
I shoot most of my photos with highlight-weighted metering, so I get a mish-mash of exposures I have to work with in post, but I don’t blow highlights. I’ll look over my batch proofs and find sequences shot under the same lighting, modify the first one to suit, then delete the other proofs and batch re-process them with the first image’s toolchain. I do this with my own software, don’t know how you’d do similar with darktable, but you might look for a way to copy white balance and tone tool from image to image…
I am using a Fuji camera, but my feeling is that JPGs are not the solution because, in the end, I edit almost every JPG anyway, so there is no significant benefit to using JPG.
Edit them in batches.
I’ve just started to do this. I usually photograph four or five musicians on stage. I use the colour ratings, one to each musician. This makes culling and editing simpler. Especially with copying and pasting settings.
I haven’t explored tags. I’ve used rating and colour. dt provides at least these means of “batching” pictures, whatever the subject.
Just use the same settings for all photos!
The only difference should be exposure adjustments and crops.
My answer to this problem is to create a style that is applied at the start of the editing process. This then gives a consistent starting point. @DanielLikesDT solution was to use the camera’s JPG, but I want the extra benefits of editing RAW files. So my style simulates the look of the camera’s JPG while still being a RAW file. But what is really noticeable is the extra detail and sharpness of this edited RAW file compared to the JPG. It is so much better.
I created this style by opening the cameras JPG in DT and taking a snapshot. I then opened the RAW file and set the exposure module to match the JPG image. I then used modules as shown to make the raw image more like the camera’s jpg.
I found I had to play with chroma values in the color zones module to get a similar color look to the camera’s JPG. White balance alone wasn’t going to achieve this.

I’ve been struggling a lot with non-JPG default looks in Darktable, but I’ve recently started to like them. So reverse engineering JPG looks is no longer the right starting point for me. For now, I plan to experiment with styles that use only one or two modules together. There’s a great YouTube video about this: https://www.youtube.com/watch?v=O4m77HAP0DA. If nothing else, it’s a fantastic way to speed up editing in Darktable. However, achieving consistent editing is still a challenge. Maybe this could be an idea for a new tutorial, @s7habo?
I rarely find anything that is universally great for a starting point. You can certainly use a set of basic settings and like Terry and others get a sort of general starting point, but I often find that usually I edit a batch of images that are similar by using one edit as the key frame. I will work on it much longer than the rest and then I simply copy that key frame style to similar images… so a sort of running styles but nothing that I save necessarily for later use in the hopes of getting the same result…
I’ve been developing “dailies” on this recent trip, reinforced my ignoring the camera JPEG as a reference render. I do use them, as the file explorer makes icons for the directory display; I use them to find discards (led to me turning off shutter activation by touching the LED screen, was getting a lot of “butt dials”).
I had one set of images where the dominant feature was snow in bright light; after looking at a few of the early renders while the batch script was still running, I killed it, deleted the tone curve and restarted it, and went back and used a control point curve for most in order to appropriately shape the large dynamic range in a way no parametric (filmic, sigmoid, etc) curve could handle. I didn’t have to handle each image individually; I’d take a sequence under the same light, do the first of the sequence, delete the proofs of the others and re-generate the proofs batch-mode with the new processing.
Essentially, ‘consistent processing’ for me is about doing such for a short sequence of captures under the same light…
As to the example contact sheet, there are significant differences in sky rendition - especially if the vacation was in one small European country.
I’m not sure what ‘style’ is as it applies to vacation shots, though.
You could ruin them all in the same way, e.g. by applying a sepia preset.
I’m obviously kidding, at least in part. In sets with wildly different scenes and light, “clean” technical processing on an image-by-image basis automatically leads to an accurately diverse-looking set of images.
You could try to create a processing profile as a preset that works for the images and the colors that appear in most of the set. Then apply that profile to all the images and go through the images one by one, carefully adjusting the preset for each image.
This should leave you with a slightly more unrealistically similar-looking set of images.
I’d also suggest getting the editing done in one session if at all possible.