I agree and I think in many cases its like a lot of these issues for many images you will never tell the difference but there will be circumstance where you might…
Taking it one step further I think to some extent your question is on point with what is going on with ART the fork of RawTherapee…its interface (RT) is jammed with multiple tools that could possibly be used for the same results and I think with ART the author is trying to simplify, in some cases tweak but in the end remove/integrate functions to streamline the software…DT could benefit from some of this too possibly
i agree with your last post too: i am an experimental guy. i use all modules and test them if they work for me and if they give me the look of my images I have in mind. i am not interested in a special look. i want to have a natural look in my pics.
if you are interested in, have a look to my last yt videos. i scanned one of my first images linear in vuescan and edit it in a handful steps in DT. Thx to aurelien pierre again and again for his work and the changes in DT.
Which probably explains both the fat paragraph and the critique.
For better or worse, many new users see that as the objective, and the job of teaching new folks that the real objective is to do much better than the camera JPEG.
FWIW, Todd, I agree with the central point of your post. Do whatever you have to do to wrangle what you want out of a given image. Don’t misinterpret someone’s advice as orders. But do be aware that the advice given is probably backed by a lot of knowledge.
@elGordo no critique was meant for anyone’s advice if so another apology…I was simply struck by the example of the question asked and re-asked re using CB for color replacement. While you could certainly do it…I think most would agree that it is easier in something like color zones (again not the only way) …I think that was actually suggested however the comments continued and my take on the whole reason for the need to use CB as a tool for that image adjustment was to “stay linear” not because it necessarily made sense or is the result of some hard and fast rule. So my intent but not likely my result was to try to point out or relax this implicitly rigid approach about only using certain module because others will mess up the linear workflow. There are certainly caveats and consideration for the use of most modules but my intent was to try to convey the idea of using the tools at hand as you see fit and not being constrained to a subset because that is the “correct” way to do it…hopefully this clarifies my post and my silly rant …it was about the notion of correctness and not anyone’s particular advice or approach…
@priort I probably didn’t express myself clearly…I may be a bit of a Covidiot myself…
I thought the main point of your post was to say that people should do whatever works to get the results they want, rather than to feel constrained by advice to use only a certain set of tools. And I agree with that. Good advice is good advice, but it shouldn’t be misconstrued as commandments.
"From M Pierre’s “Modules to be used with Care: Color Zones… Prefer Color Balance.” I take it that means you can do with Color Balance what you would normally want to do on Color Zones… otherwise it’s better to say “only Color Zones can do what it is usually used for”.
This may be a case where what is actually being inferred is… don’t use the CZ module to do overall color balancing (this could push the image all over the place)…stick to the CB module for that, however it is not actually suggesting to use the CB module to perform selective color edits to Hue Sat and Lightness as it may otherwise read…even though with some work and masks you could likely get there…
“A simple example: changing the blue sky in a photo. Perhaps small changes to its hue, its saturation, and its lightness. I know how to do this with Color Zones. But with Color Balance, I don’t even know where to start. I am trying to use it in Slope Offset Power, with HSL sliders – after all, I want to adjust H S and L of the blue. Where do I start? I thought if I picked up a colour picker and sampled the sky area… but that sure doesn’t work! I just neutralizes the sky!”
1 The point of ASC CDL is mathematical simplicity. Although this can be daunting for those of you who are used to abstract controls, learning to use lower level tools helps keep the pipeline clean and efficient.
2 Viewing color balance as a substitute of another tool is unhelpful. It has its own quirks; e.g., offset is quite sensitive, which is probably why @s7habo doesn’t use it. Each component doesn’t exactly align with shadows, mid-tones and highlights either. The headings are hints to which tones are most affected by them.
From what I can tell, the factor droppers are related to optimize luma and the hue ones are related to neutralize colors, with one exception: neutralize colors also adjusts saturation and so would be more accurate.
M. Pierre to note. (You can’t direct that comment to me.)
The article is a translation. I am willing to cut a lot of slack to the author and possible translation artefacts.
But there is no doubt that reading the article in English, the correct reading is that numerous linear RGB tools can be used in place of numerous Lab-based tools, with some kind of benefit to gain from doing so. Someone suggested in the comments above, that the benefit is only for large changes and effectively it is like using a hammer to turn a screw. If that is the case, it could have been more explicitly said. But, is it true? In his article, M. Pierre says: “.…just because the problems aren’t visible all the time doesn’t mean the problems aren’t always there. We can, up to a certain point, hide them with mathematical trickery (thresholds, opacity, etc.), but they will always end up coming out at the worst time.” and also, " the problem that arises with hue zones blending in the color zones module …which produces granular and sharp transitions." and also, “you (may) say “as long as I’m not blurring my images or working only on color, I can still use Lab”. That’s partly true, but in fact, even in those cases, working in linear RGB is simpler, with faster algorithms that can tolerate more extreme adjustments.…”
It certainly gives an impression, ñ’est ce pas?
P.S. “unhelpful”. Perhaps reconsider the choice of word.
Yes. This was about clarifying how color balance can be applied. I thought bringing in discussion about additional modules would make this more complicated.
Yes, that’s the main argument.
No. Here it is important to know that you are working in linear color space where the changes are physically correct, i.e. you avoid unwanted artifacts.
Color balance is one of the modules that work in this space. The recommendations Aurélien makes refer to using these modules as far as possible.
This means, for example, if it is possible to do the color corrections with the modules that work in this color space (like color balance, channel mixer and so on) and, if necessary, only fall back on other modules at the end of the processing. It is not meant as a complete replacement for the modules that work e.g. in LAB.
This is a very reasonable recommendation, which I can confirm with my experience in image editing with darktable.
By the way, I don’t see color balance as a rough tool at all. On the contrary, it is a very versatile module, with whom, if you know it, you can do most of the work. Here is an example video where I mainly used color balance:
dont take a reccomendation in a fundamentalistic way.
Aurelien points out, that there are modules working in a linear way, an others aren’t.
The linear modules keeps the information provided by the sensor, the other doesn’t - so he recommends using linear modules as long as possible if there are linear and nonlinear alternatives. It ends when using filmic - the most recommended tool itself terminates the linear pipe.
So care about results - if you’re missing something in the result, that was present in the original sensor data, then you removed it at a decent point of your edit.
But don’t expect information can be forced into a reduced bit depth without loosing something. You just can take over control when this happens …
I should have added an emoji it was a tongue in cheek comment…for sure given the extent of your previous post it would not have been good to add that in to the mix at this point…thanks as always for our comments…
Wasn’t meant to be disparaging. What I meant was that you may be comparing apples to oranges. The tools while similar are not the same or interchangeable. As I and others have explained, it has a role earlier in the processing pipe because it is simpler mathematically and in implementation.
What I meant was that you may be comparing apples to oranges. The tools while similar are not the same or interchangeable. As I and others have explained, it has a role earlier in the processing pipe because it is simpler mathematically and in implementation.