I'm very glad that Samsung finally allows to capture High Dynamic Range DNG files (for better color and tone control) . It was the case with Pixel phones since a long time, but it is now the case for several Galaxy phones with the "Expert RAW" app. The major improvement over recording a DNG from the Pro mode of the regular app, is that the produced DNG is not only a single exposure crappy RAW file, but that it includes the computational tricks, such as multiexposure blending and such, without excessive overcooking.
[ATTACH alt="Picture with the main "WA" module of an S22."]3132894[/ATTACH]
Picture with the main "WA" module of an S22.
[ATTACH alt="Picture with the tiny "TELE" module."]3132895[/ATTACH]
Picture with the tiny "TELE" module.
These edits are to my taste, without oversharpening, a bit of added grain and avoiding an HDR look to remain quite natural. Let me share the DNG files with deep shadows as well as strong highlights so you can try by yourself : https://drive.google.com/drive/folders/1QkPBGN5KFThJVYphM5nvz2xG1FeIOOGq?usp=sharing
To get an idea, destpite the tiny sensor, especially for the tele module, the dynamic range is pretty nice IMO.

Exposure compensation for the highlights (sky and right building). That's actually how the DNG gets opened by LR with zero exposure compensation.

Exposure compensation for the shadows. Note that I use RNI presets by default, which compress highlights in an analogue way, that's why the sky doens't totally clip.
A the end, I'm very pleased with the tonal range I can get from these tiny camera modules. It's obviously not on par with what you get from a FF sensor, but for casual yet a bit controlled pictures, it is very decent IMHO.
The drawback - of course - is that it is possible to get unpleasant artifacts due to multiexposure.

Edited to my taste. The black car in the background is severly ghosted.
[ATTACH alt="Picture with the main "WA" module of an S22."]3132894[/ATTACH]
Picture with the main "WA" module of an S22.
[ATTACH alt="Picture with the tiny "TELE" module."]3132895[/ATTACH]
Picture with the tiny "TELE" module.
These edits are to my taste, without oversharpening, a bit of added grain and avoiding an HDR look to remain quite natural. Let me share the DNG files with deep shadows as well as strong highlights so you can try by yourself : https://drive.google.com/drive/folders/1QkPBGN5KFThJVYphM5nvz2xG1FeIOOGq?usp=sharing
To get an idea, destpite the tiny sensor, especially for the tele module, the dynamic range is pretty nice IMO.

Exposure compensation for the highlights (sky and right building). That's actually how the DNG gets opened by LR with zero exposure compensation.

Exposure compensation for the shadows. Note that I use RNI presets by default, which compress highlights in an analogue way, that's why the sky doens't totally clip.
A the end, I'm very pleased with the tonal range I can get from these tiny camera modules. It's obviously not on par with what you get from a FF sensor, but for casual yet a bit controlled pictures, it is very decent IMHO.
The drawback - of course - is that it is possible to get unpleasant artifacts due to multiexposure.

Edited to my taste. The black car in the background is severly ghosted.
Attachments
Last edited:







![[PIXEL PEEPING] EXPERT RAW app. Focus is on the wall paper above the guitar. The lighter guitar string interrupts at some point because of processing and this cannot be recovered, indeed ;-) [PIXEL PEEPING] EXPERT RAW app. Focus is on the wall paper above the guitar. The lighter guitar string interrupts at some point because of processing and this cannot be recovered, indeed ;-)](https://www.dpreview.com/forums/data/attachments/2659/2659478-93f16c521d5ad715a2a64fc0f6b1e640.jpg?hash=mC3G69uH8_)
![[PIXEL PEEPING] STOCK APP raw file developed in LR (light NR, sharpening with adequate masking, texture added) [PIXEL PEEPING] STOCK APP raw file developed in LR (light NR, sharpening with adequate masking, texture added)](https://www.dpreview.com/forums/data/attachments/2659/2659484-31e58c4c932b5b321b0ee5648b09cea6.jpg?hash=sWVFHE0l3o)
















