An introduction to Local Adjustments and some worked examples are now available in English in Rawpedia: Local Adjustments - RawPedia
The labels and tooltips in the software have also been completely revised to hopefully make them clearer. Of course there is always room for improvement so if you see any errors or have any suggestions please let me know
This has been a big job and I would like to thank @Wayne_Sutton Wayne Sutton for his help with the translation and for his suggestions.
do you mean the developer version? I have the “Download RawTherapee version 5.8 released on February 4, 2020” for Windows. Probably I will have to wait a little bit, no problem, I am very very very satisfied with RT, as it is.
Do you know if there will be a new version of RT soon?
I think it is time to go to 6.0. We are now so far removed from 5.0 that the program is almost a completely different program from the user’s viewpoint. We have added dual demosaicing with capture sharpening, wavelets and now this. Put another way, if adding local adjustments on its own does not warrant a major increase in the version number what on earth would?
I notice that some of the tools have an enable fast fourrier transform (or FFTW) option. Perhaps a tooltip saying what difference that makes would be useful. The tooltips on the blur mask checkboxes just duplicate the tooltip for the title.
The easiest way I found to get the hang of what’s going on is to sharply increase say the Lightness in the Color & Light module. Then enable the L(L) mask and pull the curve down about half way at the boundary of the light & dark-gray areas. Then click through the various mask display options.
Thinking about the labels for the various mask viewing options, perhaps ‘Show modifications with mask" and "Show modifications without mask’ would be clearer if they were changed to ‘Show modified areas with mask’ etc.?
Any thoughts anyone?
Yes which is why I said that if local adjustments on their own would not qualify for a jump from 5 to 6 then what would?
That said how different does the program have to be from version 5.0 before we say it is different enough? Do we just go on with 5.10 having the next major upgrade and how do we stop people confusing it with 5.1.0
Actually it was user error (which is why I edited out the original post). I had forgotten there there three things I needed to enable to get things to work.
I am trying to use the wavelets (Local Contrast & Wavelets) and have selected wavelets. I am trying to work out how to make an adjustment to (say) wavelet level 8. It is not at all clear to me how I do so, as the controls are completely different from the controls in the wavelets tab. I have activated local adjustments, added a new spot and enabled the spot. (I made it cover the entire picture because that is what I want). I set the scope to 100 to make sure it captures all colours and textures. The mask shows everything is selected. I then added the Local Contrast & Wavelets and enabled that. (In passing I have to say that having to enable three things to get one thing to work is a sure fire way of having people forget to activate one and then post that the tool does not work).
I then select wavelet (as opposed to the default of unsharp mask) but then what? The first tool is local contrast by level. I bump up the attenuation to 2.5 to make sure I can see what I am doing and select linear. On the wavelet levels slider I set the 4 control points so that only level 7 is active. There is no change to the picture when I turn off and on local adjustments. I change to equalizer and I get a graph that somehow shows adjustments based on luminance rather than wavelet and a tooltip that might as well be written in ancient Greek. What on earth is the absicca, what is the ordinate, what does near concept luminance mean and which axis is which? Leaving the line at the default there is no adjustment. I move the line to the top keeping it horizontal. A small change but nothing like the size of the change that I get using the wavelets tab. Interestingly (to me) knocking the attenuation back down to one does not reduce the effect so I am now wondering what the attenuation does. (Also if all I am doing is keeping the line horizontal then why does linear not have any effect.)
All I want do do is mirror the effect of a global control and it is difficult to work out how to do so and even then not completely effective.
So after all that what am I saying? Two things I think.
(a) It would be helpful if the controls for any local adjustment matched the equivalent controls for the global adjustments.
(b) If adjustments are not based on luminance then the graph illustrations should not be luminance based. If they are contrast based then I suggest at the left a box one half in dark grey and the other half in light grey and at the right one half in black one half in white . (If you wanted to get pretty you could have the same as the current luminance but on the left the top half in dark grey and the bottom half in light grey and on the right the top half in black the bottom half in white with a transition). If the adjustments are based on wavelet level you could have a little wave at each end, a short wave for high frequency detail and a long wave for low frequency detail.
So that I am not just complaining without being any help, I offer a translation for the tooltip. I don’t think there is any need to add “near concept luminance” as a clarification for local contrast. I think the tip should say something like “the X axis shows the existing local contrast level and the Y axis shows the change to be made to the local contrast” (assuming my understanding of the control is correct. If that is the case (and I realize this is not a documentation point) I think the default shape for the curve should not be a horizontal line but rather that used for the normal wavelet control as shown in Wavelet Levels - RawPedia. I also wonder if there is any need to have an attenuation level if the equalizer mode is used.
