problem with sky

Tone equalizer really behaves good. When using a slightly different setting the mountains in the back get darker

How much light did the mointains in the back have? I’d assume most of them in the shadow, with the light coming somewhere from the right and slightly from the front.

Best regards, Andreas

PS: I still struggle with how to tell darktable to take the xmp files I want it to take for whatever reason… …just stopping and copying does not always work, whyever.

First one is from Slovenia (part of Karawanks mountain range).
Second one is from France (Auvergne region).

Really nice version of both photos and a promising slider in filmic module which I did not know.

1 Like

Yes it was challenging. First I set exposure on the brightest section of the image and then I chased shadow recovery options. I didn’t want to over brighten the shadows.


DSCF3221.RAF.xmp (22.1 KB)

1 Like

My edits using RT


DSCF1914_RT-2.jpg.out.pp3 (18.8 KB)


DSCF3221_RT-1.jpg.out.pp3 (15.0 KB)

1 Like

I feel like you have taken the image background way darker than it likely was. Nothing wrong with that… but I think it also dramatically changes the lighting and the way the sun coming from the right would light the scene…

1 Like

Are you using the buttons in lighttable? They’re often overlooked… :slight_smile:

Make a duplicate first if you want to keep your own edit.
image

A little apocalyptic perhaps. :volcano: :zap: Impressive.

Thanks for the images, came in handy for testing stuff. For reference here’s what I got with dcraw + gmic simple tone curve for one of them (no other tweaks):

1 Like

DSCF3221.RAF.xmp (16.3 KB)

1 Like

With these I just worked on the jpgs in gimp
Dehaze for the first one and levels auto input balance for the second


1 Like

What problems do you have with the sky? Sky in your version looks pretty good.

Exactly.

Here is my version. It’s similar to yours, with less contrast, softer textures and a little color grading to better separate green trees from orange-yellow ones:

DSCF1914.RAF.xmp (12,8 KB)

The second photo, in the given light, doesn’t make much sense to me. The road with trees as the main motif in the foreground are in deep shadow in contrast to the very strongly lit background. It would be much better the other way around.

9 Likes

Thanks Steven! No I did not find these buttons before. Where is it explained in the manual?

At least I found now how to write $(WIDTH.SENSOR)x$(HEIGHT.SENSOR) in special topics → variables in order to see a pictures original size.
Best regards, Andreas

Here: darktable 4.2 user manual - history stack
(sorry, I’ve somehow ended up in the old 4.2 manual - it’s the same though).

Out of interest, how - or in what module - are you using the variables? Just curious - I might be missing out!

darktable 4.6


DSCF1914_02.RAF.xmp (34.8 KB)


DSCF3221_01.RAF.xmp (17.4 KB)

2 Likes

DSCF3221.RAF.xmp (13.0 KB)

9 Likes

Thanks, it was a quick and dirty interpretation though, on your part you really nailed it, I mean fulfilled the request I think. You’re on a roll lately ! :smiley:

I loaded your sidecar file to take a look at how you tempered all these harsh small details / high frequency contrasts and added what I saw to my tool list ! thanks for the lesson, as always :slight_smile:

2 Likes

DSCF1914.RAF.xmp (15.3 KB)

2 Likes

With ART 1.21


DSCF1914.RAF.arp (19.1 KB)

Greetings. Roberto


DSCF3221.RAF.xmp (21.1 KB)


DSCF1914.RAF.xmp (20.5 KB)

1 Like

Your version of sky is nearly perfect, but in mine, there is still a feeling of unnaturalness. In high dynamic range photos, I encounter quite a few issues like this. I noticed you started using sigmoid instead of filmic, perhaps due to primaries? Maybe sigmoid is the solution to my problem, but I really don’t like such a big transition in changing my workflow (which is probably also a significant issue in darktable, and the reason why most users will never adopt a program that changes so rapidly). In the second picture, I see the main motif on the left, but it’s not perfect, of course :wink: