As already mentioned, for a natural look filmic will do it. A shadow is a shadow, and unless there are important things to see it can act as a border.
The LR example doesn’t look really natural to me.
I like using tone EQ and contrast EQ to add a little “mystic” touch for photos like this:
9 Likes
a nice example that “global tone mapper” looks simple, effective and natural 
… (and should not be considered “out of date” … my opinion 
)
Well, no not exactly……Having watched the video of @anon41087856 I decided to put the information about clipping warnings and filmic to a test based on the photo shown in this thread. It’s no problem to tune a nice photo a little by means of some of the many (70) tools in DT. When you are faced with a challenging photo like the one shown in this thread it’s not so easy.
Since the scene referred workflow is strongly recommended and possible to choose in the presets you are likely to use filmic in combination with exposure in every case. Is this wise or should you sometimes choose another approach?
In the scene referred workflow filmic (very often) darkens the shadows by default. I don’t understand why. Since the primary objective of filmic was to compress the dynamic range then the highlights should be darkened and the shadows lightened, so the image is compressed into the range of the output media.
You are able to adjust the sliders in filmic to lighten the shadows a lot and do many other adjustments as shown in this thread. But of course you will need other modules to obtain the best possible result.
I like the output of filmic after these adjustments. But is filmic the best tool in this case? Or put another way: when is filmic the best choice if the image doesn’t need compression of the dynamic range?
1 Like
Filmic really just let’s you control the upper and lower bounds of your image effectively. It also gives you felxability in retaining color in your highlights.
Whether the dynamic range is too much or to little, filmic is a good choice.
1 Like
In the end, it is just an s-curve that rolls off the shadows and highlights. This is why in the session, he used other modules to help flatten those areas before applying local colour contrast.
1 Like
This is something that comes up. I’ve been mentioning this since the very first implementations of filmic. Filmic is by design focussed on midtones and highlights. Its been getting better though and with tone eq you can recover shadows fairly well.
You are still fighting filmic as it crushes black by default. But using filmic it seems to be something to get used to. Sorting it out is a second step but can be done well.
Previously people have said, yes avoid filmic when the assumptions behind it arent satisfied. Now it seems many find that it can and should be used all the time.
You’re wasting time if you expect an „if this then that“ approach for darktable. There’s no such thing.
Theres more a juridical "it depends on the case“.
You can use filmic to shoot yourself in your foot and you can use it to improve your edits especially if your‘re brightening already bright highlights as a collateral effect on tweaking midtones or shadows. You can achieve similar or even better results using other modules, but also not in an „if this then that“ set of rules.
my other version looks strange on my monitor. i think i need a new laptop. 
I think filmic (costum option) does a pretty good job on this picture with this brutally harsh light. Filmic looks the most natural to me here.
translated by…
Another look on shadows and filmic_DSC_2481_01.NEF.xmp (8.5 KB)
1 Like
No you are right! The LR edit doesn’t look natural and the result is far from what the scene looked in real life.
I like your edit:+1:.
Just to clarify: the phrase “shooting yourself in the foot” in the video doesn’t refer to filmic but to do editing, using any tool and to putting too much emphasis on clipping indicators. Furthermore the thresholds in clipping indicators the are set way too conservative by default.
How would you edit the image?
Since i don’t need the poppy lighroom look for me filmic incl giving more room for shadows + tone equalizer to improve shadows a bit ist sufficient for this image in my opinion - others have other also valid opinions.
I’m using filmic by default - just because i can adjust midtones via tone equalizer or exposure and do further stuff without bothering about clipping highlights.
Of course, this is a quite lazy approach but why fiddling around with tone curves, multiple masked exposure corrections to bring highlights back into a proper range when filmic does this for me…
But if the default workflow isn’t sufficient, then I’m not stuck to it - using darktable is about having options 
1 Like
After working the image in rawproc a few ways, I’m pretty sure the “poppy lightroom” look is some form of color saturation, as the recorded data doesn’t come close to those colors in any tone operation.
To the overall topic: Thing is, filmic is just another tone curve. Make a synthetic image 1000x1 pixels in dimension, and tone each pixel gray with the value of its X coordinate. Run that image through filmic, gamma, reinhart, or any other tone operator, and the resulting 1000x1 image can be used to plot the curve of the tone operator. Look at those plots for a bit, and you start to get a sense of what they’re doing at all tone levels.
Anything called ‘filmic’ is about a toe at the low end, some sort of linear midsection, and a gradual rolloff at the high end. That toe, if flat, will crush shadows, but it doesn’t have to be shaped so extremely. That’s what @anon41087856 is endeavoring to give you control of, how flat the toe…
1 Like
Don’t forget we’re dealing with a number of contradictory requirements:
- you want to compress your camera dynamic range (12-14 bits) into your screen or paper dynamic range (8 bits),
- with a decent global contrast,
- and keeping a nice looking local contrast in all regions of the image.
In other words: you want to compress the global dynamic range, while leaving local dynamic range largely untouched…
So something has to give, and often lack of details in the shadows is less objectionable than lack of details in the highlights.
In addition, the default setting now is to add 0.5-1 EV of exposure. indicating that cameras tend to under-expose (understandable, given the ugly look of overexposed highlights). But that means that the shadows suffer even more.
But this is every curve ever applied to a photo?* It’s the first thing one does learning photo editing.
The curious thing with the darktable implementation is how difficult it is to shape said curve because of certain assumptions that don’t match photos like this one very well. ART, RT and Photoflow implements the tonemapping aspect and leave the curve to a curve as we know it allowing it to be easily shaped to avoid crushing the black to much. I know filmic does things beyond and differently that these but it’s still a useful example imho.
So simple automatic log tonemapping with strength at 53 and the following curve
Gives this result in a matter of seconds. (I’m just trying to have more detail in the foreground portal whilst preserving detail in the building not doing anything else)
For your typical ETTR architecture shooter a lot of photos have these characteristics so my curves are very often tiny toe, biiig soft falloff at the highlights. It can get way more exagerated than this example with high dynamic range cameras.
DSC_2481.NEF.pp3 (15.1 KB)
my bold
But the thing is one would expect a new fancy tonemapping module to give quick and precise control over those tradeoffs. That often above is as I see it, but I have absolutely no understanding of the math or programming, an unnecessary assumption that fairly frequently is wrong.
*gross over simplification but good enough to be correct
The difference is how far to the left is the toe. Base curves I’ve seen finish the dip almost a quarter into the histogram. The original Duiker filmic curve when plotted in full in normal display resolutions doesn’t show the toe, you have to zoom in on the lower tenth or so to start to see it.
Yes that’s no surprise that they have different shape. Interestingly my criticisms of dt filmic is that the toe does what you claim the Duiker doesn’t. I find it difficult to make the toe small enough while the falloff large enough using filmic. You have to use other tools to get there which again is sort of strange considering the tonemapping goal of filmic.
Where you prefer your crushing to be has nothing to do with math, math comes into play deciding how to crush your shadows or burn your whites.
As far as this image is concerned, first thing I did was add 2EV exposure (so 2.7EV total) to “fill the histogram”. So this image was certainly not ETTR looking at the raw data (causing part of the noise problems in the shadows).
However, when applying the Nikon basecurve, just correcting the in-camera exposure compensation is already too much… And let’s not talk about what happens to the shadows…
Afaik, that basecurve mimics the result of the default in-camera processing (as represented by the embedded jpg).
1 Like
Sorry I wasn’t communicating clearly I was referring to my photos not this example.
No, the sooc curve toe goes further in. Its a custom curve with extra small toe. The file has also been tonenapped using a log module. Thats why I thought it was a good comparison to filmic. Similar, but not identical, modules. With the normal more flexible curve you can do without local editing.