The same modification but with the second file GWRrailways42.RW2 instead of GWRrailways43.RW2
I commit a change, with
- replace “int” by “double” for “deltaE weakening” “Exposure expcomp”, “Sharpening Radius and Blur radius”
- to harmonize with main menu, and also more accurate
You must clear cache (or delete cache)
jacques
1 Like
Thanks for the pp3 Jacques, I look forward to trying it out. I guess the best strategy is to try and adjust highlights with other tools which have more image information to play with, then lighten areas of the image using locallab.
@spidermonkey,
I use Local Lab’s Tone curve to take care of that.
Included is a pp3. I hope it helps.
@jdc,
Salut Jacques, thanks for your hard work and the fun time. I wish you the best with this wonderful project. Merci 
I will be away from RT’s Local Lab.
there are other ways to get there, for example
GWR railway 42.RW2jdc3.pp3 (19.2 KB)
The first spot, to reduce the exposure of the sky, is an exclude spot, on the half upper image
The second spot, to increase the exposure of the passenger and railways is a normal spot
1 Like
We can improve the result, with 2 others spot, to to lighten the shadows in the trees
GWR railway 42.RW2jdc4.pp3 (26.7 KB)
To compare with original
I just added “shadows Highlight” (only in Lab mode). This allows a good improvement in the case above
I just commit today (March 4th)
- 2 GUI changes : 1:1 to some expanders and Display hue chroma luma ref to mask curves (thanks to Pandagrapher
- add slider to improve mask with “GuideFilter”
Jacques
2 Likes
Used last version of locallab with good results in a short time, i’m very happy with it.
I hope to see localab in official release of RT.
839 crashes when I try to export a picture.
W10 Pro
Last few versions work well for me, no crashes for months.
RawTherapee_newlocallab_5.5-847
https://keybase.pub/gaaned92/RTW64NightlyBuilds/
Is running fine here.
Did also try “Spot removal tool” 5.5-256 -----
But it crashed when I tried to open an image.
Unfortunately, RawTherapee_newlocallab_5.5-847 crashes quite often.
Here is an image and pp3 that always crashes on opening - opens momentarily, then RawTherapee shuts down.
Running on Windows 10
GWR railway 42.RW2 (14.8 MB)
GWR railway 42.RW2.pp3 (23.3 KB)
(Yes that picture again, I have taken others too!)
Tried clearing the cache, still a problem
@spidermonkey
Effectivly it crash… whenever, I desabled all “rubrics” as “exposure”…
The 3d spot generate a crash,even in this case it’s empty…no action…
If I delete, and add a 3d spot, no crash…curious…
jacques
Thanks for looking at this. Like you say, curious!
I know this thread is a very long one and I have not read all the relevant entries. So what I say might have been said before!
I wanted to see what the normal version will do to this photo. I am happy that using just the two modules (Highlight recovery with Color Propagation and a negative Graduated Density filter) I got the following result. What do you all think?
GWR railway 42.jpg.out.pp3 (11.6 KB)
2 Likes
I have also found that RawTherapee_newlocallab_5.5-847 doesn’t always save the processing profile - seems to be set to default the next time you open the image. It definitely seems to have a problem with images that have had edits made with locallab in a previous version
That one is nice! The sky is well preserved. I would perhaps tone the shadow recovery down a little bit, to make it look a bit less washed down, or increase local contrast in this area.
1 Like
@spidermonkey Is it possible to have an example of pic/.pp3 files for which you encountered this problem? I don’t understand why it happens: when using a .pp3 file created by an old version, only missing parameters are normally replaced by default ones
Thanks,
Pandagrapher
In fact we have search why ?
@Pandagrapher ask me to verify “range” of “exposure slider”
2 mounth ago, I change “integer” range -200 + 400 (which were divided by 100) by “double” range -2 +4.0 (not divided)
In your file values are about 270 or 400, thats about 100 times more…In consequence crash in some cases
As we are in development, it is not possible to predict this case
So do not use old pp3 with values that will be out
Thank you
jacques
Thanks for the explanation, that makes sense.