I recently had a set of high contrast B&W negatives scanned. I adjusted them to taste and then saved the resulting high contrast images. Both input & output files are jpgs. The resultant files were mostly OK, but a couple had their shadow detail replaced by a nasty noisy grey. See below.
Hi @Jon! What adjustments did you make to these images? The difference could be caused by one of the tools.
I used dynamic range compression, but other images with it were fine. I adjusted contrast, but not greatly. Those were the only adjustments that both pictures had.
The set of negs generally had some very deep, but not entirely empty shadows. Most worked out OK.
I just tried reprocessing without the dynamic range compression and the nasty noisy grey areas disappeared - shame the picture was so dark though 
If you are able to share one of the images… others might be able to make an alternative suggestion for your edits to help avoid this and perhaps enhance the output…
Below is one of the original images, but down sized. - Having discovered that dynamic range compression resulted in dud output I tweaked a tone curve & used a bit of shadow adjustment, which gave a similar effect to the one I was after, but not quite what I really wanted.
I’ve seen weird shadow noise with the Dynamic Range Compression tool, but they were visible in the preview. It’s a tricky tool to use. It is sensitive to individual pixels, so the preview is only accurate at 1:1. It’s also non-local, so the preview is only accurate if the full image is available. Put these two together and it means the preview is only accurate if it’s showing the full resolution image. That’s unfortunate because the tool is very good at compressing high contrast scenes while maintaining a natural look.
Can you show how the preview looks like so we can see what result you are after?
You’re right - the noise / greyness is fully visible at 100%. I’ve got acceptable versions of my two photos so, as the they aren’t my greatest ever pic’s I’m happy to leave them be.
Although the discussion has been helpful to me the main reason that I posted was to flag up a bug in rawtherapee.
It is difficult to give an informed opinion without the files. In addition, JPGs have a lower dynamic range than Raw, at least 1 to 2Ev.
I made a comparison of the various methods for processing high dynamic images:
https://rawpedia.rawtherapee.com/Local_Adjustments#An_evaluation_of_the_dynamic-range_capabilities_of_tools_in_the_%E2%80%9Clacam16%E2%80%9D_development_branch
There is also a comparison (a little older) on shadow processing
https://rawpedia.rawtherapee.com/Local_Adjustments#Five_ways_to_change_the_exposure_and_lift_the_shadows
The latest version of Dynamic Range & Exposure (last commit) is a little more efficient (especially the Local Adjustments version, with a Saturation control checkbox), but it is an aggressive tool.
Jacques
The original files were jpg scans of HP5 B&W film processed normally, so the original negatives have a relatively limited dynamic range (which I was pushed to its limit.) - I uploaded one of the original two problematic files without any further processing above. Both of my dud processed files can be seen at the top of this thread.
Thanks for the links, I’ll definitely check them out. Many of my digitally captured Raw files, processed to give 16-bit tiffs, are of extremely high contrast subjects. - Going digital has given me access to subjects that I just couldn’t handle using analogue.
1 Like