This is the work of a former "darktable" developer (Aurélien Pierre) who decided to fork the codebase and go it alone. He has strong opinions about the correct way to do things. I like some of the cleanup on the UI that he has done. For now, Ansel and darktable are compatible in terms of the underlying database. So you can easily switch back and forth between them. If the fork diverges significantly, it would be more difficult to maintain the compatibility.
darktable has seen some major changes in the past few years, moving away from a "display-referred" to a "scene-referred" workflow. Aurélien contributed a lot of code to make that work, most significantly, the Filmic module. darktable is not as user friendly and as polished as other commercial tools (Lightroom, Affinity, Capture One) but it is capable if you take the time to learn it.
Aurelian's youtube channel is pretty insightful. He explains a lot of what is behind the scene referred modules. And as you say, he has been driving a lot of that work. I've been using Darktable for many years for all my photo editing and it has improved massively in the last few years. He does a great job explaining how raw file processing works (in any raw photo processing tool) and how Darktable does it.
Having come across his rants on Darktable last year, I think he does have a point on the UX front. A lot of the filtering in the light table (where you organize your photos) is a bit of a mess of confusing buttons, options, and weird convoluted abstractions. I never liked that part of the software. I can work with it but years in, it remains counter intuitive how you get your files in there and manage them. It's just convoluted and weird all over. A lot of bad ideas layered on top of each other. Aurelian kind of freaked out when last year some pretty major changes were just pushed through without much debate. And I agree with him, it didn't really improve much things.
Anyway the magic is in the darkroom part, which is the part where you edit photos. There is a wide variety of modules aimed at different expertise levels. Scene referred mode is basically a big upgrade over what a lot of other packages do, which is to blindly apply pre-defined curves for cameras without much regard for the actual pixels in the raw image. Filmic and other modules do this a bit more intelligently by actually looking at the pixels, using some heuristics and working from the lowest levels of the pipeline all the way up to do the right things. It adds up to a lot less work when editing photos for me compared to earlier versions. Mostly photos come out pretty OK without much tweaking. I might tweak perceptive saturation a bit, add some contrast in filmic, etc.
Basically, the workflow is roughly: 1) tweak the exposure as needed for the gray point. Filmic adapts with sane settings for black and white points and you typically don't have to tweak that. 2) add some contrast in filmic 3) maybe add some local contrast 4) in the color balance module fiddle with perceptive saturation. Done. There are a few more things I do for sharpening, profiled denoise (as needed), etc. But that's pretty much it. One nice thing is being able to apply defaults based on rules to photos.
I might play with Ansel a bit if I can find some time over Christmas. Kind of curious to see what he's done to lighttable and the rest.
I'm not familiar with the software, but generally, I get why he left to go do his own thing. I've made a few forks of a few projects over the years considering it.
I'm perpetually in the process of beating this dead horse, but FOSS would have so so many banging gold-standard user-facing apps if they enlisted the help of experienced UI or maybe UX designers and really worked to make them part of the community. To do their job right, they need to talk to the community to figure out what their needs are, and if maintainers shrug their shoulders while the few people who speak up are skeptically bikeshedding everything they say into oblivion, then we're also going to shrug our shoulders and walk away. That's what happened to me the several times I tried to contribute to various FOSS projects as a(n experienced, professional) designer rather than a(n experienced, professional, former) developer. Often, the response you get for merely intimating that something could work better if it was set up differently is like calling someone's kid ugly.
It would be like a team of civil engineers working on a restaurant design scoffing at an architect that specializes in restaurants offering to help make an effective kitchen layout. From the civil engineers' perspective, the architect's input is superfluous and would probably slow down progress. Meanwhile, everyone else that has to interact with that kitchen suffers.
I agree with you point about UX-professionals. I am not a professional UX designer, but I work in product/graphic/web design since 20 years. As such I volunteered for an open source project once, in a push which (according to the founding developer) was of great value to the project. Basically I just looked at all things as if I had seen them the first time and tried to formulate solutions that replaces gotchas with discoverability, makes unexpected things expected, sand down paper cuts, etc.
This only worked because the founding developer was the benevolent dictator for life and I had one guy that I needed to convince. And that guy clearly was a genius in what he did and accepted that I was better than him in what I did.
Now I don't know the Darktable project's organizational structure, but given the grievances aired here I assume there is no clear shared vision and nobody feels responsible for being really in charge of the software as a tool that solves problems.
Now I am a nerd myself, but there is a kind of open source nerddom, where the people are in it for the coding first and not primarily for creating an elegant and nice to use tool. This can be okay, if there is someone in a deciding position of the project that at least cares about that aspect. If all contributers are just fiddling away in their own corners of the software you will get a patchwork of a software where different parts feel completely different.
I think that a part of FOSS culture for some developers is a backlash against the things that bother them when coding at work, and a designer having more say over how the interface works than the developer really pisses a lot of people off.
> where the people are in it for the coding first and not primarily
Oh man, you are not wrong there. The amount of pull requests and contributions I have seen that basically amount to a bit of refactoring for the sake of refactoring in FOSS certainly is higher than in non FOSS environments. Which likely has to do with there being more checks and balances in corporate software development.
I don't think i've ever agreed more with someone on here. I'm really not sure if its a case of UX/UI and art people not being willing or not seeing the demand in the FOSS community or FOSS communities to seek out those people to participate but dear god so many good FOSS projects just have horrid UI/UX because the guys doing these projects are great at backend but horrid at frontend, things like amazing AI apps just thrown in streamlit and called a day drive me nuts.
I've thought a lot about this. I think it goes much deeper than people merely not being great at interface work-- I think a lot of FOSS development is a backlash to the sort of development people have to do at work, where they're forced to reckon with designers that have more say about what the interface looks like than they do. Also, in many instances, these developers tell themselves that they are making great interfaces but they're just a little bit ugly, and people need to read the docs and get over it. Most computer users who use dozens of application every week will never read a single complete paragraph of software documentation in their entire lives. Why? Because they don't have to-- and many of these FOSS applications are doing things a lot simpler than what MS Outlook does.
Also, anybody who's been to art school or mentored junior developers knows that we get most defensive about the things we're least confident in. Unfortunately, that comes out when trying to address UI problems in FOSS projects.
> I think a lot of FOSS development is a backlash to the sort of development people have to do at work, where they're forced to reckon with designers that have more say about what the interface looks like than they do.
I think you've captured the crux of the issue and the same applies to designers too.
Say I just spent the last eight hours trying to convince a stubborn front-end developer that the reason customers complain about our UI and the product keeps failing usability tests is that it needs a UI overhaul and that telling people to "just RTFM" will not win us any new contracts.
Getting home and having a similar debate with a FOSS dev team where I have even less sway does not sound like a rewarding way to spend my free time.
> It would be like a team of civil engineers working on a restaurant design scoffing at an architect […] Meanwhile, everyone else that has to interact with that kitchen suffers.
> FOSS would have so so many banging gold-standard user-facing apps if they enlisted the help of experienced UI or maybe UX designers .
And here lies the rub. I have tried soliciting help in the past . Most ui UX ppl only want to work on successful projects, but successful projects don't need help from ux/UI people.
A) If you surveyed the same number of developers to contribute to the project and they said no, would you make the same inference about all developers? I think you're assuming more than you realize.
B) The number of successful FOSS projects with interfaces good enough for people who don't have a working mental model of the way software operates is vanishingly small. Firefox... though they actually have a formidable team of designers. Inkscape I'd say. But almost every successful FOSS project caught hold in the technical community, and no further. Sometimes it's barely good enough for that... I mean hell... look at Eclipse. Compare its features on paper to what you get in commercial editors and then see how many developers use it voluntarily.
C) One thing few developers understand about interface design is that adding features here and there to make things better doesn't really work like it does with, say, an API. It involves analysis, talking to people who use the software, coming up with a strategy, and implementing that strategy. Usually that strategy isn't the sort of thing you can implement piecemeal, which is why most significant UI updates today involve making a completely new design, and letting users enable the entire thing as they see fit.
> If you surveyed the same number of developers to contribute to the project and they said no, would you make the same inference about all developers? I think
This was not an assumption. You are stating a fact. I simply can't find as many UI UX ppl who are willing to work on open source..
Those I have asked reply with the answer above.
I don't know what b and c have to do with what I said though.
This is what i kinda figured to be the case, UI/UX and generally artists don't seem to participate in projects as much for free or for projects that aren't already successful where they'll get some publicity out of especially on the art side of things.
And this is one of the big problems... not what you're talking about, but your understanding of what we do.
'Art' and UI/UX design are as different as fiction writers and technical writing. Someone might be good at both, but they're definitely not the same thing. Interfaces are a communication medium, and reasoning about the best way to communicate something is a process that often doesn't even touch aesthetics. These types of designers working for larger companies probably don't even get invited to the meetings where aesthetic decisions get made, and they definitely don't work for the art director who'd be involved in that.
The first step is figuring out what problems your users are trying to solve, and the next step is working with them to figure out the most effective way to do that. It's pretty pointless when the users are insanely defensive about the status quo, as is the case in most FOSS projects.
Many UX/UI design folks I know have unpaid side projects, they're just not FOSS.
Nonprofit org websites, event posters, flyers, t-shirts, illustrations, small utility apps, WordPress themes, unsolicited redesigns of well-known applications on Dribble, etc.
I spend my days convincing developers to make UI changes. Spending my nights and weekends doing the same thing but with even less authority does not sound like fun.
Planning on it. I'm compiling a list of points and counterpoints I've encountered when discussing this over the years and forming it into something informative that has practical actionable advice for everyone involved.
On the flip side, coming from Darktable and trying out Lightroom felt like trying to do good old darkroom work with my hands tied behind my back.
Darktable offers a few really powerful, generic tools that you can use in different ways to get different effects – things like equaliser, parametric masks, LAB curves, etc. It makes little sense to use it without reading up on some of those more advanced tools first.
Lightroom, in contrast, focuses more on offering a small selection of pre-defined tools for specific purposes. But once you want to do something outside of that (parametric masking is one of those things I really missed) you're shit out of luck.
You are most likely both right, because it mirrors my experience everytime I look at Photoshop/Lightroom when my dad does things and when looks at darktable when I do things.
Would you mind giving examples where you use parametric masking? I read the docs* and it is not very clear to me the practical side (I'm a Lightroom user).
An example: Camera noise is often particularly annoying in the sky area. So I'll have two denoising filters active: A subtle one that doesn't lose much detail, and a strong one that only applies to the sky. Selecting the sky tends to be easy with a parametric mask: Just select the sky's hue and it's done.
It’s not stellar, but it doesn’t suck, and I haven’t been forced to upgrade.
Maybe I will never migrate off LR and just create a new catalog on some other computer with Ansel or something.
I use the last version of LR6 one could purchase outright, and I dragged it kicking and screaming all the way into Ventura, but I’m sure the party well end pretty soon.
But as a Darktable user, I found the implementations of strong opinions frustrating because I value actual workflow above potential technical superiority.
Display referred worked fine for me because the important work happens before the shutter is clicked. The skill I want to develop is fixing things in the lens not fixing them in post.
Breaking changes suck.
But again open source developers don’t owe me anything.
I don't disagree, I spend a lot of effort thinking about things at exposure-time. I think the thing that sold me on scene-referred is that I started conciously thinking about the kind of data I was capturing. It's often the case that the display medium, or its artistic representation, has a much lower dynamic range than the sensor in the camera, and so there are a number of things you can do to get the most data possible when capturing to leave yourself room for expression at presentation time.
What this gives you isn't necessarily the ability to "fix" things in post, but the ability to decide how you want your image presented in a certain medium or format when "developing". Even master photogs like Ansel would take liberties when developing to realize their vision from the negatives they were able to capture.
For me, I have intuitions of the relationship between my printer and my screen from experience.
And I run test prints and reprint if I don’t like the way it prints.
I mean since you mentioned Adams, for Adams the print was what mattered…the print is the title of the last of his three books series.
Kodak doesn’t change HC-110 every six months because doing so destroys value.
But again it’s his software. I just wish he weren’t so bored as to invent problems to cleverly solve and solutions for which he may argue.
The legacy editing workflows are generally, if not entirely, still available in Darktable. I haven't yet encountered a defect in my old edits on new versions of Darktable.
It absolutely is, it is what I use for RAW developing and post. Does everything, does it good and is, above all else, non-destructive.
It is, at least of I go by my dad who is a Photoshop veteran, quite different to what he used to.
The only things I could see as, if I am really really critical, are:
- sharpening, usually decent enough but SharpenAI from Topaz is another league. As I said, good enough for all except the edge cases
- de-noising, same as sharpening, astro-denoise works like a charm so
- masks, which I haven't really figure out yet, so my problem and not darktable's, Lightroom and Photoshop seem a tad more intuitive so
One thing I'd love, but again maybe I just didn't find it yet, is the possibility to export the database of picture edits. For now, I have darktable create dedicated files once a photonis edited and I make back-ups of those. I did loose edits for around 100ish photos due to some unrelated issues which required reinstallation of darktable so, which again was on me for missing darktable stopped creating said files after an update. Being to just dump the darktable database of edit data would be really nice so.
Summary, I can only recommend it. Especially for people starting post processing, the learing curve for each software is basically the same after all, and without prior knowledge they wont recognize any differences.
> "Ansel is what Darktable 4.0 could have been if its developers were not so busy turning it into an usability nightmare. Ansel is a Darktable 4.0 variant where 30.000 lines of poorly-written code and half-broken features have been removed, and 11.000 lines rewritten : it runs faster, smoother, uses less power and requires less configuration. Enjoy an app focusing on getting work done and stability."
This opinion comes from a former pho photographer, one who desperately wanted to love Darktable...the shots are warranted in my opinion. Darktable is a complete car crash of poor usability and militant contrarianism on established user experience design.
It has so much promise but the 'lead by committee' approach just resulted in some kind of collective 'demand avoidance' from the devs who seem to revel in delivering an unusable product. And I mean unusable for those not willing to learn an entirely new paradigm of interacting with a piece of photographic software and dive into the docs for everything, including stuff as silly as a keyboard shortcut or move between modules.
I've yet to read all the Ansel blurb but I'm pretty sure this is from the guy who's made the most improvements to Darktable in recent releases. So it's incredibly exciting to see.
I doubt it'll get any DAM capability though, even for a fork that is asking too much :-)
As a darktable user, I have to say none of that would motivate me to switch.
But then I care about the results, a nice picture to print or look at on a screen, way way more than about the tools used to get that result. Goes for cameras and lenses and whatnot as well.
> I have given 4 years of my life to the Darktable project, only to see it destroyed by clueless geeks playing code stashing on their spare time, everyone pushing his own agenda with no sense of design, in a project where nobody is responsible for anything and where we work too fast on everything at the same time.
Anyone know what happened to the Darktable project? I've only used it a few times, but it seems nice for someone who knows how to use it (which isn't me!) but curious what drama happened.
When I switched to Linux full time about a year ago, darktable was a godsend. I currently have it pointed at a NAS share with ~8k raw images and it works pretty flawlessly and I can barely tell that the images are offhost.
Personally, I really like a number of the features that this fork removed (like the timeline). The software does have a bit more of a learning curve than the commercial variants out there, but darktable is impressively capable and I personally jive more with its theory-based approach.
I _do_ agree that deprecating the display-referred mutations is a good thing, but the darktable documentation is already pretty adamant about scene-referred being the future; though there's no way people are going to go through their entire library and update their past tweaks from display to scene-referred, so I don't know that those modules can ever be removed.
I guess if I was going to be as petty as the author of this fork, I'd say that the Ansel logo looks awful.
The author is the "Dr. Rant" (his own words) of Darktable, so a lot of what he says is probably exaggerated. A lot of it isn't, though - Darktable does indeed lack focus and direction lately, leading to unnecessary bloat and unresolved issues. In other words, typical FOSS problems that sometimes lead to a fork.
This is just pure excess cleverness. It's not even obvious how many times this loop runs for. There's a comment 1000 lines away that says this:
#define NUM_INSTANCES 5 // or 3, but change char relative[] = "-2" to "-1"
This is just insanely unmaintainable code. If a string is warranted I would have just done the "stupid" thing with a `sprintf(buf, "%+d", relative);` to make it obviously correct, even if it seems "slower".
I had to think about what that loop does for an embarrassingly long time.
I’ve never seen XOR used as an in-place alternation operator like that in a for loop. I’ve only ever seen it used as a swap function not an inline expression.
I’ve also never seen a for loop that uses a mutable string updated char-by-char like that.
> However you put it, there is no valid reason for a software left open without touching it to turn the computer into a toaster, especially since we don’t buy Russian gas anymore.
Darktable is still going strong. I am a pro photographer and use it. I tried Ansel but most of the useful modules are removed and personally, I really don't care much for the "right way" of doing things from an engineering perspective. Darktable works and that's why I use it.
darktable is still being developed. There are pretty regular releases. I use it as my main photo processing software (raw-based workflow).
As I said in my other comment, Aurélien has strong opinions. He felt he could not effectively work with the other darktable devs and decided to fork the project.
That's kind of the nature of non-destructive editing. While I do think the operation stack could be smarter about collapsing edits that are commutative, once I got a workflow order down I haven't really worried about this much.
The documentation actually has a pretty good example workflow that flows through mutations in an order that makes sense for most cases. Though it's not really that simple, and so you'll find yourself trying to relearn each time if it's something you do only on occasion.
It’s not your fault. It has the worst UX of anything I’ve used. Be careful if you use the mousewheel anywhere, you don’t know what unexpected action it’s going to perform. Drawing vector shapes for masks is wildly frustrating too. It does have a lot of powerful modules though… and it’s free.
I actually would prefer it to zoom, which it sometimes does. A couple of examples: if you're using the tone curve tool it'll change the exposure for the level under the pointer; or if you're hovering over a vector it will scale the vector when all you want to do is zoom on the vector so you can move its fiddly control handles; but then if you're hovering over one of the settings dials by accident it will move the slider left to right. Absolutely something I wouldn't expect and have not seen in any other software. The UX is mind-boggling inconsistent and unpredictable.
If you're using nix, I was able to make a build from the darktable derivation[0] modified to use fetchgit (since he hasn't made a github Release since December) and `s/darktable/ansel/g`.
I use Darktable quite extensively for underwater photos, but keen to try Ansel out and see if it's less friction.
An example of friction in darktable:
- I have an external strobe which means I have to put the exposure down to its lowest when shooting, otherwise everything is washed out. Darktable in newer versions has "Compensate camera exposure" on by default, which washes out all the images until I click it off. I'm sure there's a way to make this checkbox disabled by default, but why can't it accept what comes out of the camera?
- No favourites: there used to be a way to have all your panels in a favourites tab. This was great as I usually only use a handful of modules that I use. It's gone in later versions
- The "color balance" panel, not to be confused with "color balance rgb", it's not in any of the default tabs but useful for saturation adjustments. Why are some of these useful modules hidden? Shouldn't all modules be available by default. The only way you can get to it is by searching.
- White balance: there are now two modules and it warns you if you adjust one or the other: "white balance" the standard one on the "base" tab and "color calibration" tucked away on the "color" tab. Both modules are turned on by default, but if you adjust one or the other without turning one off it has a big red warning.
- One upgrade decided to reset export settings, and so my EXIF data was stripped out when exporting. It took me way too long to figure it out.
You can create a preset for the exposure module and define rules when it kicks in. For example based on the camera manufacturer, focal length, iso, etc. I use that to increase the default exposure compensation with my Fuji. I deliberately underexposes to prevent highlights from clipping. So I usually want a +1.25 exposure compensation. Likewise, you might want denoising on for high iso files.
You can organize modules into profiles and simply hide all the ones you don't use. The default profile hides some of the deprecated or display referred modules. You can change this.
White balance indeed has a deprecated variant for display referred and a scene referred one that works completely different that you typically use together with color calibration (which is where you should do most of your color correction, including color temperature changes). The reasons are mathematical and beyond me to explain properly (Aurelian does a great job on his Youtube channel). It boils down to not throwing away the baby with the bath water in terms of rounding errors accumulating and switching color model (to the one used by your display) too early in the pipeline. It might look pleasing but then it bites you when you want to tweak tone or do other things. This is the whole point of working with the scene referred modules.
Having all the legacy modules around is indeed somewhat confusing and Aurelian solves this in Ansel by hiding all the deprecated modules now. They are there for legacy files still.
My favorite photo development program remains lightzone, I find it has by far the most intuitive work flow that let's you achieve extremely advanced steps in a quite straight forward way (much of the functionality can be achieved in darktable but it's often very unclear how). I originally bought a copy but it has since been open sourced. Unfortunately it's photo management features are really subpar and development has been quite slow.
I wish I could use it just for the development and use something else for the management, but last time I tried that wasn't really possible. I have looked into contributing several times, but it's written in java and I really don't the have time to invest in getting up to speed first.
Somewhere very recently I encountered a quote from Ansel Adams, in which he seemed to anticipate that "electronic" photos would be the next big thing, in (IIRC) 1980. Wish I could find that again. Seemed remarkably prescient.
“There’s no end in sight. Electronic photography will soon be superior to anything we have now. The first advance will be the exploration of existing negatives. I believe the electronic process will enhance them. I could get superior prints from my negatives using electronics.
“Then the time will come when you will be able to make the entire photograph electronically. With the extremely high resolution and enormous control you can get from electronics, the results will be fantastic. I wish I were young again!”
I have only read The Negative, but from it I get the impression that Ansel Adams is really good at boiling things down to fundamentals and understanding the key concepts involved in things (much like I gather Elon Musk can be, I suppose). I'm not surprised he was able to grasp the benefits of digital photography also.
Electronic imaging in terms of television has been around since the 1940s or 50s. We all watched Neil Armstrong walk on the moon. It was a pretty safe bet that resolution would get better until it exceeded film.
Kodak had more patents in digital imaging than any other company. Why they failed is partly because they underestimated how much the overall photography market would decline and what needed to be done to be successful in the 10% that was left.
The winners were electronics companies (notably Sony and Panasonic). The losers were the photographic companies who were really chemical companies. Fujifilm was one of the few that survived because they went into (amongst other things) cosmetics and coating films for LCD screens, both of which leveraged expertise they already.
In early 2000 a lot of photographers were saying that digital photography is not "real", it doesn't have a "soul", it should not be allowed in photo contests and it's just worthless in general.
To be fair, early-2000s digital cameras were ... not great. 4 megapixels if you're lucky (not that megapixels are the only thing that matters, but they do matter, up to at least 20 megapixels).
Looks good, I'll give it a whirl. I could never figure out a good darktable workflow. The UI seemed all over the place with some basic features missing and way too bloated in certain areas that I imagine 99% of users would never touch. Hopefully Ansel has figured out a better feature balance and UI.
My current Linux photo processing tool is Another RawTherapee [1], which is a wonderful blend of the power of RawTherapee with a UI that has a lot of similarities to Lightroom.
I also throught of ART as a fork of RawTherapee, also forked with the aim to make it easier to use.
I wondered if that is something hard to avoid: Over time features get added, usability worsens. This might be general, but it could apply to FOSS in particular, which often takes a modular approach (cue UNIX philosophy) and rarely includes structures where potentially useful features are rejected in order to keep the whole experience simple.
Yeah, I guess there's a point where starting afresh is beneficial to refocus. A bit different, but it's the same for institutions too. I know some countries have "sundowner" clauses on every institution, where they need to apply for a permit to continue operating every 10 years or so. Provides a good point to think "is this working and still serving its original purpose?".
I can't get this to build for MacOS, has anyone succeeded? I ran the brew commands from the mac-nightly CI script, did `git submodule init; git submodule update` and get this error running `build.sh`:
```
$ sh build.sh --build-type Release --install --sudo --clean-all;
In file included from /Library/Developer/CommandLineTools/SDKs/MacOSX14.0.sdk/usr/include/cups/http.h:39:
/Library/Developer/CommandLineTools/SDKs/MacOSX14.0.sdk/usr/include/netinet/ip.h:189:2: error: unknown type name 'u_char'
u_char ipt_code; /* IPOPT_TS */
Did you see the OS Support section in the README [0]?
I tried for about 15 minutes and didn't get it to build. And even if it did, it would probably not be stable enough to use for anything productively.
> On the following picture, I made the styling, the make-up, the lighting, the shot, the editing, the retouching, the software color filters, the documentation to use them, the website to talk about them in 2 languages, and even the colorspace used for saturation adjustment. You will find very few people with this kind of full-stack understanding of light and color able to also write efficient computer programs and read academic research papers on applied mathematics. Yet, you will find a lot of image editing applications and a lot of guys trying…
Jeez, wish I had 1/10th of the confidence of this guy
Weird, I scrolled down to see what cameras it supports, and there was a link labelled "Supported Cameras" that goes to https://rawspeed.org/CameraSupport.html but that's a squatted domain now. Is there a list somewhere? It doesn't instill much confidence in me if I can't even see what camera RAW formats are supported.
I for one am glad that someone with a more holistic approach takes this on, even if Aurélien is really snarky. Darktable really is a usability mess and is easy to use incorrectly.
I'm also glad he is taking donations, since the darktable project won't take any. It gives me hope that this would give him the freedom in actually implementing his vision.
Agreed. Darktable has always been an impressive project, produced great results, great features and the UI was IMHO above average for an open source project. But it did suffer from lack of holistic view, many features were kind of duplicated across modules and it was unclear what's the best way to do things. It seemed like people had a lot of freedom to implement their own thing in modules.
I think Aurélien's snarkiness is quite distasteful and might prevent getting some of them on board. But he has a proven track record and I believe he can execute his vision, so I will be donating as well.
About the Ansel fork, I've used Darktable for 6 years and tried Ansel the last two weeks: if you have never used Darktable, Ansel UX is better for a first time user imo, more streamlined. Keep in mind that I'm biased as I agree with most of the author's points here: https://ansel.photos/en/news/darktable-dans-le-mur-au-ralent...
innovation in raw developing software is always very welcome. my main gripe with basically all of them is that the controls to manipulate colors are very tricky to use for the precise work that color grading requires.
"Curves" is still the de facto standard even if it is a usability nightmare, and apparently simple stuff like "take the non saturated, light blues and move them towards cyan" is very hard or straight up impossible in many raw converters. This is especially painful if you compare that to video grading software and its companion LUT building software, where new paradigms to manipulate color has emerged.
I sincerely hope that this is an effort that will bring the state of the art forward - the author's stance on scene referred workflows is a good premise
This might actually get me to check out Ansel as a current Lightroom Classic user. I tried Darktable a year or two ago, and there were some immediate dealbreakers. I really like the idea of someone capable and opinionated forking a popular project. I’m rooting for him!
Out of curiosity what’s keeping you on Lightroom classic. I moved my admittedly amateur photography collection to the “new” Lightroom a couple years ago and have found it to be an improvement in pretty much every way
- UI is snappier
- Integrating syncing lets me leave my laptop at home when travelling and taking photos.
> You will find very few people with this kind of full-stack understanding of light and color able to also write efficient computer programs and read academic research papers on applied mathematics. Yet, you will find a lot of image editing applications and a lot of guys trying…
Wow, I'll have what he's having for confidence.
On the other hand, open-source photography and bitmap editing is still waiting for its own Blender or Godot and I applaud anyone willing to have a go at it. What's available (GIMP, RawTherapee, darktable, ...) can mostly sort of get the job done, but if you're the kind of person who seeks relaxation, pleasure and aesthetics in photography, the open-source software feels just too geeky, too unfocused, too unrefined.
I'm currently on DxO + Affinity and even though they cost quite a lot of money, I'd probably shoot and enjoy the whole thing significantly less if I had darktable and GIMP waiting at home.
It's pretty common with open source projects. The people with time, energy, and an "itch to scratch" and the people who really know an area aren't disjoint sets, but people who fit in both are rare. Lot's of projects like e.g. GIMP start off making mistakes that would have been obvious to someone who understood the domain well, but the team makes them because they are learning as they go.
Very much so. That, and he's French. The French way of critiquing is a bit different from what other cultures do and can be a bit abrasive. Cultural differences are a thing to be aware off when dealing with people from different nationalities.
He means well and is deeply passionate about this stuff. He's worked his ass off in Darktable for many years. One of the few (only?) developers that was trying to do this full time on donations, etc. He's the closest thing Darktable had to a benevolent dictator style leader (like Linus Torvalds). Big loss for the project to not have him around. I hope they can reconcile their differences somehow. Would be best for all.
Basically, what happened is somebody pushed through something that in Aurelian's mind were some severely misguided, low quality changes without much debate or process. It just showed up and he flipped out and did not appreciate being bypassed/ignored like happened.
Nikon "High-efficiency RAW" support is missing. This is IMHO the fault of Nikon and their vendor TicoRAW. If you're going to come up with a new RAW format, then it is your responsibility to commit the decoder to open-source libraries! Sure, patent and license the encoder, but if you keep the decoder closed-source and proprietary, then you're a <insert expletive>.
The installer is an EXE instead of an MSI. Publish an MSI! Use the "wix" tool in your build pipeline, it's not that complex.
The installer and the deployed app are not digitally signed, which throws up a litany of scary warnings and errors. Other open-source developers have gone to the effort of signing their builds.
On first launch the app flashed a command-prompt window up and then disappeared.
The "Exposure" tool has an automatic setting to compensate for the camera offset. Okay... then why did I bother to offset the exposure just to have ansel undo my intent by default? Then... it layers on a default +.7 EV for like no reason...
The overall GUI is completely non-standard and bizarre. I've never seen anything even vaguely like most of the UI controls in this application anywhere else... ever. It's like someone who's never seen a GUI in their entire lives invented everything from scratch.
The non-standard UI elements like the combo boxes aren't just weird, they're buggy too: they don't work consistently with the mouse. I can see the option being highlighted, I can click it... and then it'll pick something else! They only work reliably with the keyboard. In general the "selected item highlight" appears to be off-by-one, but not consistently. It's bizarre.
Colour management is a mess. On a HDR OLED monitor, Lightroom can display wide-gamut and HDR images correctly if the desktop is set to HDR. E.g.: if toggle an image from SDR mode to HDR mode then the only difference is that highlights get brighter and some extreme colours become more saturated. Everything is correct by default and SDR tones don't "shift". In ansel, the colours look wrong and any setting I choose in the menu makes them even more incorrect.
There's a lot of talk of HDR support on the site, and the menus even "suggest" that PQ/HLG support is there... but not really. The export formats are all from the 1990s and modern HDR formats like HEIC, AV1F, JPEG XL, etc... are missing in action.
A lot of the options/alternatives seem like developers being too reluctant to get rid of old, bad code. For example, if a new superior demosaicing algorithm is added, then it's usually best to just drop the worse ones! I tried every option available, and all but one was broken, to the point of returning super-green results or just all-black. Lightroom for example just uses a really good demosaicing algorithm and doesn't burden users with a choice of a bunch of bad alternatives.
In general, the same criticisms apply to ansel that applied to Darktable: over-complicated, too many options, most of which are either wrong or useless. Internal details exposed to the end-user in the UI that should be debug traces for developers, not permanent GUI elements. Easy to inadvertently "break" the whole thing by reordering pipeline elements by dragging and dropping something accidentally.
I suspect that on top of the ~30K lines of code Aurélien Pierre deleted from Darktable 4.0, another 60K should be deleted...
Interesting philosophy, but just tried it with "bokeh" and well, the results are pretty bad? "Related keywords" are interesting, but many miss the last letter. The result list shows some reviews for old smartphones. Nothing relevant at all. Whereas if I type this into Google, results look way better.
Tried the reverse: One related keyword for "bokeh" is "out-of-focus". I typed "out of focus" into both Google and Chantal. Google: first result mentions bokeh. Chantal: Also mentioned somewhere, but not as a related keyword.
Nice choice of keyword. I kind of suspect "dumbrish" needs to be more of a first pass for classifying things as opposed to the whole story.
Interestingly, using it as an exact keyword seems to improve some of the links but just empties the related keywords list.
This feels like an approach that can work really well for topic-specific searches (despite the hole you spotted) but that probably won't ever generalize. Still... I'd have liked to have it for a couple of topics that I used to patrol forums and post basic answers for.
I’ve used Darktable and would happily try Ansel. But in recent years ios compatibility became an important feature for me.
My camera can send raw files to my phone wirelessly, with an ios app I can edit them on site and have a draft ready instantly. This is especially useful for events or anything time sensitive - but convenient in other use cases as well.
Still, I have good memories of Darktable and the author’s filmic module, I hope to try Ansel anyway at some point.
For personal photos, integration with the iOS Photos library is something I'd like to see from more apps. RAW Power is pretty much the only raw editor that integrates well, without needing to copy files around and creating duplicates.
Agreed. iOS integration is really key for me. But I can understand that the developer doesn’t have an Apple device and isn’t interested in personally supporting it because of the cost and Apple’s attitude towards open source.
lmao, Aurélien Pierre has been making insanely big leaps in darktable, putting it leagues above its competition and providing something so unique compared to Lightroom and others, so I'm not surprised that he just created his own fork.
So much pent up anger - and I understand! Darktable has a strange approach to evolving their software. They somehow keep everything old and bad as they add improved versions and you end up with tons of ways to do very similar things, and to add to this mess, Darktable often uses its own nomenclature, or a very very technical one. If you're coming from Lightroom or Capture One, you feel like taking your first steps on a new planet. For no reason!
So, I'm sold! I'll definitely take a look at this. He gets it!
Oof. Looking at the "Transitioning from Darktable" page ( https://ansel.photos/en/doc/special-topics/from-darktable/ ), the author is extremely disparaging about Darktable developers. Very off-putting, even if some of the changes seem sensible.
[edit] Make that quite a lot of the changes, actually, including getting rid of some really dangerously bad bits of UI design. Still dislike the guy's attitude but I've got to admit he has a point.
This guy basically changed darktable from a shitty Lightroom clone to something truly unique very recently, and his approach is highly divisive among the users.
Yeah, having read the article linked above [1] I see his point - it definitely looks like he's put in an enormous amount of work for the project, and I can see where his frustration comes from!
To me, Darktable has a 'feel' of a highly technical editor for people who really care about colourspaces etc, and although I hardly know anything about colour I rather appreciate that. It's interesting to note that almost all the modules that contribute to that 'feel' are written by Aurélien.
Beautiful open-source photo editing app, and that's really saying something. A tremendous achievement. However, the lack of a macOS build shows a fundamental lack of market knowledge & project management skill that will ultimately doom this project. It always surprises me when technical acumen is combined with complete blindness to a significant potential customer demographic. I hope they wake up.
EDIT: Am a professional photographer who frequently interacts with other pros on assignments and their own art projects. We're discussing tools constantly and have no qualms between paying for commercial tools or paying in the form of dedication to ascend the learning curve of a less friendly tool. We'll figure out a tool if it's worth it.
> Mac OS support is not active, the main reason is only 4% of the Darktable user base runs it while the continuous integration bot for Mac OS needs weekly maintenance (for breaking unexpectedly). It may or may not work, but if it doesn't, I don't own a Mac box to have a look, so don't expect anything. Mac OS is anyway not playing nice with Open-Source, having deprecated OpenCL and all.
“Ansel is what Darktable 4.0 could have been if its developers were not so busy turning it into an usability nightmare. Ansel is a Darktable 4.0 variant where 30.000 lines of poorly-written code and half-broken features have been removed, and 11.000 lines rewritten : it runs faster, smoother, uses less power and requires less configuration. Enjoy an app focusing on getting work done and stability.”
Definitely not a well known product name, as someone who plays PC games and also does some CUDA development, I've never heard of this product from Nvidia.
I never heard of it. After looking it up, I don't get the appeal. Not sure why someone would want to take fancy screenshots of video games unless they're marketing a game.
Nothing gets my blood boiling like a Steam/App Store game listing that only has these "in-engine but not the actual gameplay" screenshots. If it's a FPS, show me the screen with the gun and the crosshairs. Show me the HUD in a driving game, the position of the camera in a 3rd person RPG. I don't care how some random NPC or a building looks from a position I'll never get the camera to when playing the game.
Photography, just like in the real world, but with lovingly-sculpted virtual worlds and (often) the ability to pose the actors in the scene as you wish. Dead End Thrills [0] is a master of this, both artistically and commercially, but much of his prior content seems to be down, which is a real shame.
Maybe Pierre can pull it off himself. He’s clearly knowledgeable and has strong convictions and Darktable is incredibly powerful, so I’m hoping the good bits from either camp can be reconciled into something lasting, minding the bus factor.
Darktable’s issues and lack of coherent stewardship is quite noticeable.
darktable has seen some major changes in the past few years, moving away from a "display-referred" to a "scene-referred" workflow. Aurélien contributed a lot of code to make that work, most significantly, the Filmic module. darktable is not as user friendly and as polished as other commercial tools (Lightroom, Affinity, Capture One) but it is capable if you take the time to learn it.