Is everything the same with mask tools as before? About selective editing tab. How can I highlight a face in a photo with these idiotic masks?
It’s impossible to use. Well, really impossible. It’s terribly inconvenient.
Easy. Open a photo, go to Selective editing, click Add, then a spot appears. Drag the spot to a face and change its size if needed. Then add a tool to this spot, eg. Color & Light, then drag sliders to highlight the face.
Please ask your questions without being insulting. Next time I’m just going to remove the post.
I suppose the use of the word “idiotic” was uncalled for. But he is expressing a level of frustration with this tool that many of us feel.
(And I assume “really” should have been “nearly”).
Are these masks in RT…which I have only used a little to be fair not really a version of the U-points used in NIK and then DXO and maybe there was a version of the Nikon raw software that used them…so it’s not some half baked made up weakness of RT just a different approach that might have also have been the best or easiest option to integrate into the code.
Perhaps you missed the worked examples in RawPedia? Local Adjustments - RawPedia
no, but the principle is similar and anyone familiar with U-points should not have any trouble with RT-Spots, especially in Basic mode. The main difference is that you have far more control over the detection parameters in RawTherapee compared to the DxO version.
https://rawpedia.rawtherapee.com/Local_Adjustments#Introduction
Years go by, and nothing changes. Only the tone equalizer was added. I don’t take into account the film negative and the like. In reality, the only color apperance/lighting tool that can be used from curves is. 4th tab. If we are talking about photographs for people. No. There is a certain established philosophy here. Please excuse me. It is impenetrable.
Indeed as with most things RT lots of options to fine tune things… I just meant “conceptually” similar…
Yes you are free to express your frustration, but you need to do it without being insulting or derogatory. Maybe, one could even offer some constructive criticism.
At any rate, what I’ve said is clear.
I have spent a fair bit of time with those tutorials, but still find it very challenging to apply to my own photos. When I have a bit of time I’m planning to submit some “unworkable” examples here to see if we can figure it out what I’m missing.
As regards the selections tools of RawTherapee I find them very powerful. I am also a DXO Photolab user (on version 8, now) since many years now. I chose to buy this commercial software precisely because it was similar to RawTherapee
This being said, if you read the forums of DxO (which I do…), there are quite often users asking for much powerful options to select objects.
Right now users are spoiled with commercial softwares like Lightroom or Capture one with their super powerful ways to select objects thanks to the AI back-end (which indeed works like magic…)
In the end, it is nobody’s fault.
I have read, over and over, on this forum, that the RawTherapee developers are not contrary to improve them (quite the contrary, actuallly…).
Currently, they “only” lack the time or the skills (in essence, GTK stuff) to do so; needless to say, on their spare, very limited, time (for free).
@anon43125638
Right now, IMHO, you are probably better off to work with ART [1], if you want “better” selections tools (e.g. more similar to GIMP).
From what I always read on this forum, darktable, which is also open source to run, has very powerful masks tools. Therefore, you might try it as well.
for to create a mask, you used 19 clicks, in RT you only need 5:
it’s a different approach, but after once learned, you will love it!
I use rt not because of powerful tools, but simply because it suits me. It’s like you don’t have to have a Nikon z9 or a d850 to take good shots. Each person has their own.
For example, you can use Davinci Resolve or 3dlut creator to develop photos. Yes, you can get a quality result, but only a quality one and nothing more. You should like the photo, no matter how many mistakes it contains. I can also use art. But I miss the saturation curve for lightness in that program.
Well, so you isolated a certain color in the mask.
Everything is simple here. For me, RT is a good developer. Masks, retouching and other things are not its strong point. All this can be done in Photoshop, exporting the image to a tiff file.
By the way, as for skin tone, I haven’t seen anything better than rawcolor app. It’s free. True, it’s already closedsource. In other programs, it’s like a dusty bag was thrown over the person. acr, c1, dt, rt (sorry). Rawtherapee is certainly better, but it’s a hodgepodge of different approaches. Like, it’s cooler this way. Everything is the same as … only in profile.
This is not comparable to any “dxo”.
Hello everyone and excuse my bad English.
@anon43125638 , I am the designer of these “idiotic masks”. Should I take it to mean “village idiot” or “useful idiot”, or ??. I prefer to think of it as a stylistic device to show your dismay. You feel the same way about the general philosophy of the software, which you find impenetrable.
Thank you all for expressing your point of view and a certain form of support. I appreciate it.
Just a historical reminder, 5 or 6 years ago the number of developers (what’s a developer?) was at least 5 or 6. Today, for various reasons, there are only 2 people (a little more) who do most of the development. Among them, there’s me, a sickly old man of 77, who’s absolutely not a computer scientist and is quite incapable of making certain modifications, particularly those affecting the GUI.
I admire the abilities of others, such as :
- Alberto’s work @agriggio integrating CTLs into ART is remarkable - I’m not talking about the CTLs themselves.
- @Lawrence37 work on (almost) real-time updating of executables, either from current “Dev” or from Pull Requests, is a substantial plus.
- Aurélien Pierre’s creativity in Darktable is also worth mentioning.
I’m not doing RT development to flatter my ego, nor for any kind of report, nor for glory, but for the most part out of personal challenge. It’s by examining RT and seeing other products in action that I decide whether or not (depending on my computer skills) to create a new algorithm.
These creations include (but are not limited to) either the work of researchers (Ciecam, Wavelets, etc.), or a totally personal creation (White Balance Auto Temperature Correlation).
Selective Editing is a special case. When I bought my first D200 in 2006, I was seduced by the U-point approach, different from masks. Of course, I didn’t have the code, so I just imagined an algorithm. It took about 10 years, with a lot of help from Pierre Cabrera (GUI) @Pandagrapher and Ingo Weirich (code optimization). Depending on your habits, you may find it easy or difficult to approach.
Depending on one’s skills, habits and beliefs, software will initially be easy or difficult to use. Some people’s opinions evolve over time. For example, if there are 3 software products A, B and C. A few years ago, A was perceived as more complete than B and less complex than C. Later, the opinion was reversed: B became the best and A the most complex…yet little has changed in terms of code and algorithms. But why should they all be the same and similar to paying customers? Diversity is richness.
Now to the masks. In principle, they’re almost contrary to the RT-spots philosophy. Nevertheless, in response to demand, I’ve added some. It’s a lot of work for a mixed result. I could have copied the masks that already existed in RT, or Darktable, but I did it differently. There was a challenge with 6 difficult photos in the spring of 2024… nobody used masks, but that doesn’t mean anything.
That’s not to say that they can’t be improved - either by integrating curves and polygons, or by changing the concept a little (ART type), but it’s a big job, and one that I don’t know how to do for the most part.
One of the current challenges is to make an algorithm designed for astro-photography work for more traditional photography. The algorithm is complex (very) - I’m not questioning it, but I’m having to deal with what is rare in astro-photography, the problem of highlights. You can follow this development (Generalized Hyperbolic Stretch), here.
ghs
Gamut compression (ACES) is also an interesting point that I think will soon come to fruition.
Gamut
I’m pleased with the current cooperation at RT, in terms of code - fortunately I’m not alone, our cooperation is working well, forum opinions and documentation (thanks to Wayne @Wayne_Sutton ). There’s still a lot of work to be done:
- improve the code and algorithms: new contributors are welcome.
- improve documentation and tutorials: RT is sorely lacking in videos, and progress is being made on Rawpedia. However, documentation (content and support) won’t happen on its own, so thanks to all the initiatives underway, and here again, contributors are most welcome.
Jacques
Maybe I missed something. Is there an option in this program to check against a reference picture that I decide to add. The basics. All sorts of hue, lab curves and hundreds of other super tools - this is secondary, as for me. Having a reference picture in front of your eyes is important. Of course, you can switch between program windows to look at the reference picture, but this is a photo development program
Thank you for all of your work on this!
I had a very good experience with a RT spot last night. It perfectly masked a blue shirt on my subject, perfectly managing some semi-transparent patches. It only took a few moments, and gave me results that would have taken a lot of pain-staking work with brushes. I had a minor breakthrough in understanding the relationship between spot size, transition, and scope for this one.
That’s progress for me!
I still struggle with other applications. The documentation is good, but overwhelming (so much information!). Hopefully when I understand it better myself I can help with that.