I’ve written a thing about ART vs Lightroom, focused on raw processing. As I was writing about ART in my workflow, and thinking about why I find Lightroom so annoying, I realised that LR only suits a very specific set of photographers. And using ART has helped me see raw processing differently.
Please do give it a read. And I’m all ears for your reactions.
https://www.chuckhenrich.com/free-lightroom-alternative-art/
Nice write up, Thanks!
What I’m wondering:
When I ditched lightroom I went straight into Darktable and I love it. But how does ART compare to Darktable?
Are they different tools for the same job?
Thank you! Yes, ART and Darktable are different tools for the same job. Whichever works best for you is an excellent choice.
Personally, I find ART a bit easier to understand and work with. In particular, Darktable’s masking is challenging for me, while masking is one of ART’s strong points.
The Darktable reliance on a database is not my preference (as I mention in that blog post). But that’s because I don’t work with huge numbers of images. Commercial/professional photographers who shoot hundreds of images per session, hundreds of thousands over their career, really do need a database/catalog to manage their content.
If you’re curious, give ART a look-see. But if you’re happy with Darktable, stick with it!
Nice write up indeed, it can help people better understanding what ART is.
You may want to correct a couple things.
First, the processing pipeline is not really shared with RawTherapee’s, as Alberto says on the website the pipeline has been significantly restructured.
Second, upon first install ART applies a default processing profile IIRC, which includes applying the white balance as-shot, and applying a ton curve that tries to match the embedded thumbnail’s tonality.
Thanks for the blog post, well written! But I think a more fitting comparison would have been ART vs Adobe Camera Raw, which is like ART a pure raw converter.
Thanks Sébastien, good points!
Can you point me to the info about the default processing profile? I haven’t been able to find it.
I’ve updated the post to clarify the differences a bit:
No, but they’re related. ART was forked from RawTherapee and shares some of its technology. However, ART makes important enhancements to the processing pipeline, much better local editing tools, improved metadata management, easy-ish extensibility with external tools, and more (see the ART Features page for details).
Interesting idea @geni1105. They’re not really the same thing though. Adobe Camera Raw is a plug-in, not a standalone application. It’s launched as a component of Photoshop, After Effects, or Bridge. ART is an independent application.
Most (all?) of the points I made about the costs of Lightroom’s automations pertain to Camera Raw’s masking/selections.
Since only a specific subset of photographers will gain much benefit from Lightroom’s database/catalog, ACR in PS, AF, or Bridge is a more natural workflow: you don’t have to import anything, just open the file and get going with edits saved in sidecars.
It’s just what I recall from ART’s behaviour on first install. I have to check that.
Well I didn’t find info, but I just reset ART by removing the config folder, and opened a couple of raw files from a folder devoid of any previous .arp sidecars and I can confirm that in that case, ART applies an auto-matched tone curve based on the thumbnail embedded in the raw when it is opened for the first time.
I can confirm - by default ART will use the AutoMatched Curve enabled.
Many thanks @sguyader and @tankist02 for pointing me to this important correction. After investigating, it turns out that the default adjustments raw editors apply makes the case for ART v Lightroom even more strongly.
I’ve updated the post to insert a new section:
The cost of letting Adobe decide
Every raw editor applies some processing when you open a file for the first time. That first rendering is your starting point for everything that follows. If you accept it without understanding it, you’re proceeding on assumptions someone else made for you.
Lightroom applies the same default rendering to every raw file regardless of how it was shot. A grey overcast afternoon and harsh midday sun get the same starting point. So does a quiet portrait and a high-contrast landscape. Reasonable on average means wrong, slightly, for almost everything specific.
ART’s defaults are deliberately chosen and fully visible. The initial tone curve is derived from the JPEG that your camera embedded in the raw file. That embedded JPEG reflects your chosen picture style or colour profile. So ART’s starting point is your camera’s own interpretation of the shot, not a third party’s opinion of what looks good.
The sharpening and noise reduction defaults that ART applies are ISO-aware from the start. Low ISO files get capture sharpening. Higher ISO files get noise reduction instead. A base ISO landscape and a high ISO interior shot are meaningfully different files. They shouldn’t get identical processing before you’ve even looked at them.
All of this is open to inspection. ART’s default profiles are plain text files. The rules that determine which profile applies to which file are listed plainly in Preferences. You can read them, change them, and understand exactly what’s happening before you touch a single slider.
Lightroom also applies automatic adjustments on import. Lightroom’s default profile, Adobe Color, is a generic baseline that Adobe’s Julianne Kost describes as “designed to be a great starting point for any image”. According to Kost, Adobe Color adds warmth to reds, yellows and oranges, and applies a small contrast boost. That may suit some images. It’s a poor fit for a winter landscape, a grey urban scene, or anything with a cool palette. It isn’t a rendering derived from your specific camera’s picture style or colour profile. The resulting slider values are visible in the Develop module, but the reasoning behind them isn’t. You see what Adobe decided. You don’t see why.
Lightroom does offer something more sophisticated: ISO adaptive presets, which interpolate sharpening and noise reduction continuously across the full ISO range of your camera. Once set up, they’re more granular than ART’s three fixed ISO bands. They require creating reference images, configuring presets, and digging into preferences people never open. The feature exists. The default doesn’t use it. Most photographers never set them up.
Adobe’s defaults don’t necessarily produce bad results. Often they’re fine or even good. But they’re one-size-fits-all and Lightroom users generally don’t question them. Adobe makes the decisions, the results are what they are, and the user starts adjusting based on that baseline.
But acceptable and appropriate aren’t the same thing. When I begin working a raw file, I want the starting point as close to capture as possible. That’s the whole point of shooting raw. I don’t want developers and product managers at some corporation assuming I don’t know what I’m doing and imposing their assumptions on my work.
So I much prefer ART’s deliberately minimal defaults: conservative sharpening at low ISOs, conservative noise reduction at higher ones, a tone curve matched to your camera rather than Adobe’s preference. They clear the deck and leave the creative decisions to you. Lightroom’s defaults are opinionated. They ask you to trust Adobe’s taste. Most users do, without realising they had a choice.
ART has an advantage that, if you set up AI masking according to the instructions on the webste, you can buy Lightroom courses and they still mostly apply, except for some minor details. This is not the case for Darktable.
Pet peeve: I have not yet found a good-enough Darktable course, free or paid. And ART does not need a dedicated course, exactly because good Lightroom courses fit.
Here “good” means a course that not only explains slider mechanics, but also artistic decisions involving goal-oriented series of operations on multiple photos pre-selected so that they need particular edits. As a fair comparison target, let me note that the “Wildlife Photo Processing Essentials” course by Simon d’Entremont includes 18 raw images to practice with, covering topics like accentuating the subject, enhancing 3D spatial cues created by the light, or a tough situation with haze. I am not saying that it is a perfect course - e.g., it does not include any practice on making two photos look consistent with each other, but 18 examples means a lot of meaningful practice.
I was able to practice along with the explanations in Lightroom and then repeat the same edits in ART. Of course, the result was that I edited photos with ART, but using a workflow heavily reliant on AI masking, as Lightroom users do.
My learning path with Darktable so far consists of Standard Workflow - darktable.info – Mastering Darktable and a course by @JasonTheBirder, which, sadly, at that time, only included one photo to practice with. Now it has two.
The major unblocker from that Darktable course was that I learned to work with non-AI masks, and this cross-pollinated my work with ART. I now feel that I use ART in a more “native” way (with rough brush or color similarity mask, properly feathered to follow object edges) and thus avoid some AI mask artifacts.
Of course, I tried Darktable on my own photos, too.
Then, I participated in the Wildlife Light Lab by Chiara Talia. It’s a photography course that covers various natural-light scenarios, proper camera settings, composition tips, and editing technique. In the course, there was no requirement to use Lightroom, so I used ART on some photos and Darktable on others. The course included feedback, and the procedure involved sending both the RAW file and the edited JPEG to the teacher.
Result: no major critique on ART-based edits, and consistent “why did you decide to lose the nice colors that you had in the RAW file” or “this looks washed out” comments on Darktable-based edits. The lesson for me was that the initial bleak rendering affects my decisions on further edits (i.e., I see a pale sky and assume, wrongly from the viewpoint of a Lightroom user, that it was pale and has to be pale), so I created a preset that mimics my camera’s default JPEG rendering, to use as a starting point.
And this is exactly the point:
And this is not necessarily bad. I am not mature enough to see the correct result behind Darktable’s default rendering, and it shows, so I’d better get a different starting point.
[ART] doesn’t do layers, compositing, or retouching
False: there is a “Spot Removal” tool in ART, which can be used as a form of retouching.
I understand what you mean, but in my vocabulary, “retouching” is different from removing dust spots.
It comes down to semantics but the local adjustments are very layer-like and function in that manner: Each is a separate adjustment and all come together to create the total edit. Each can be turned off and off, with variable opacity. They function as a nominally transparent stack.
From my perspective as an amateur who mostly shoots family travels ART is faster and simpler to get good results compared to DT. ART gives decent starting point with AutoMatched curve or DCP profiles. For extra adventurous it supports integration with Spektrafilm and Spectral Film Simulations and a whole bunch of CTL scripts for various tasks. Usually I prefer colors and tones from ART compared to DT. For example, I haven’t seen such a problem as “salmon sunset or sunflower” in ART.
I’m planning on updating the course with a few more practice examples. However, just for the record, I think it’s a better idea for people to practice more with their own photography and try to understand the concepts on their own photography rather than using practice Raw files. Just from an artistic perspective, I think it can actually be harmful to edit too many other Raw files from other photographers, simply because one was not there to experience the scene.
I think a more holistic approach is superior, where one goes out and experiments. I think the practice-based approach of pre-defined problems can be more useful in other domains such as mathematics, but I think it can be more harmful in photography. I think an overreliance on technical exercises in photography can be more confusing than helpful. I understand that some people can learn that way to some extent, but in the end ultimately I think it engenders a mindset of being too boxed in to tecnhnical details.
Even in mathematics, as a former teacher, I’ve observed that too many technical and pre-defined problems can be problematic for many students, whereas (guided) open-ended exploration often leads to more creativity in the end. Speaking as a professional photogapher as well, I really think that too much focus on specific examples can stunt creativity in the long term.
I get that you don’t agree with that, especially from the 2/5 star rating you gave to my course, but I stand by it and I think that for those who really want dozens of examples to practice with, then I don’t think my course is ideal for them - sort of like how some people really like some textbooks and others need something else.
Vehemently agree. Of course, it’s important to read the documention (RTFM) for any software. And demos can be really helpful in explaing how something works. But it’s only with practice and experimentation that I learn how hardware and software work for me.
Unstructured “play” is how we all learned at the beginning of our lives (at least back in the last century). It’s still an excellent learning mode.