Beware though that it tends to not abstract away a lot of technicalities, if you dig deep enough you may encounter exotic terms like “illuminant”, “demosaicing method”, “green equilibration”, “CAM16”, “PU”, “nit” and so on, but I personally love it for that even while I am still learning what half of it all means.
I’d say the only major lacking feature of RT is support for HDR output, which hopefully will be coming by way of PNG v3 and Rec. 2100 support.
RawTherapee is better than Darktable. But that’s a pretty low bar to clear. There are reasons people pay for Lightroom.
I threw darktable and rawtherapee on the table but without technical grit you get nowhere really fast.
It's no my wheelhouse so they are mostly in there own.
Those tools you really need for properly edit raws are hidden in blated features (multiple demosaic algorithms) or completely missing (AI masking). And UI is not user friendly.
I have not found an equivalent mechanism in RawTherapee. Does anyone know if it has an equivalent tool?
Would really like to be able to use RawTherapee's dual-illuminant DCPs (not available in darktable).
I know it's a different space, but as a counterexample, FabFilter makes audio plugins that are the gold standard for that kind of interface and it isn't even close. Anybody making an interface for interacting with points on a curve should sit down with the free demo of FabFilters Pro-Q3 for just a few minutes to experience what's actually possible and how it should feel.
That claim does not match my experience in any way.
For example, Control Cage curves have node value adjustment to 1/1000th.
https://rawpedia.rawtherapee.com/General_Comments_About_Some...
Language pedantry has nothing to to with photographic image processing expertise and if anything this would be a sign that the developers care more about being "right" than what users want.
Just installed it on my m1 mac and opened a folder of RAW files. The initial loading lagged my whole macbook. Couldn't even open the dock. Once the thumbnails all loaded it's better but not as buttery smooth as I would have hoped! Would love to know what other commercial apps do that make them not lag. Is it just that they're written natively?
Already having a workflow for AI based subject masking is def nice to see.
That said, they're all GIFs and each ~10-22MB. Making loading the readme larger than the program size itself. Embedding some video would be snappier.
I'm not sure what the perfect solution is, but it is hard to sync a ton of shadow files to cloud storage, versus one big catalog file.
Is the metadata in an open format, so I can take the edits to other programs?
I am glad there's alternatives to having to shell out for Light Room every month. I only need to edit RAW files after holidays!
We need an easy to use RAW editor. For a long time I used Darktable, with default settings I would get images that where close to the camera jpeg. I just had to change in what artistic direction I wanted to go. With update after update I had to fight to even get decent skin colors.
Currently on a pirated copy of CaptureOne, but would rather use something open source (Or buy something affordable)
Do you have default camera and lens profiles build in?
I'm no AI fanboy, but it's neat to see some dreams come true because of it.
Few reasons.
1. It's 100$ a year, which isn't pocket change when it's making you no revenue.
2. Apple likes to randomly deny developer accounts.
3. They have no issue with outright rejecting apps with vague reasoning.
4. Plenty of high quality raw editing apps already exist for OSX.
If someone really wants to use it on OSX you've provided clear instructions.
And then it's sending these thumbnails back from rust to javascript as base64 encoded strings, not using a shared buffer: https://github.com/CyberTimon/RapidRAW/blob/fc21ede729b45d97...
This is the sorta stuff that native apps mostly don't do. They don't base64 an image just to send it to a different app (react) to base64 decode it (via a third app, webkit) via a slow ipc mechanism (tauri) from itself to itself, allocating 6x the chunks of memory along the way for one bit of data (the 6x are: raw data in rust, base64 data in rust, json encoded base64 in rust for tauri ipc, json encoded base64 in javascript, base64 in javascript, raw image data in webkit to finally view).
Its a high priority to optimize the loading speed of large folders and you can expect an improvement in the coming days.
Kind regards, Timon
Another thing I noticed with RawTherapee after LR was the sloooowness. Lightroom has been super optimized to show all edits (or their preview versions anyway) in real time.
DT has a highlight reconstruction module too but RTs was superior every time I tried.
They're talking about the Filmic module which is a fancy exposure adjustment curve. I ported it to Rust for use in our internal 2D compositing renderer (not open, unfortunately). That's when I got to appreciate the work done there, really.
The gainforge crate has another port, although not with all parameters of the original algorithm, AFAIR.
The algorithm is explained in a blog post by the author of the module. [1]
Who is btw. the same guy that got fed up with DT dev and forked it to Ansel.
[1] https://eng.aurelienpierre.com/2018/11/filmic-darktable-and-...
Please explain why photographers need 20 differnet sharpening methods, 5 demosaicing algorithms, many colour corrections that are almost useles if AI masking is not present?
Coders often lost in all kind of geeky features that missing actual usability by targeted audience. Bloated software is not what I would expect from alternative to commercially used proprietary software.
You can ignore features that aren't made for you, and actually I think they're mostly hidden by default in DT (make a preset if you don't like the default tool selection). All these features were added because somebody needed them at some point, the DT/RT/ART communities are chaotic and lack vision but they're actually using their stuff.
>Coders
As I said, this is not software made by coders for coders. This is exactly how the software made by photographers would look if they lacked organization, focus, and UX skills. If it was made by coders (and UI designers), it would probably have looked like Lightroom and had AI selection.
It would be great if RT supported something along the lines of “take the entire canvas after initial raw processing but before CLUT and final touches and put it into a temporary file; allow me to do something with it in any other tool; then load it back and apply the rest of the steps on top of that”, but that might require it to store that edited canvas somewhere in addition to .pp3 file.
>immensely grateful for Google's Gemini
>AI Studio's free tier
Be sure to take a close look at Nitro, created by a former Apple lead of Apple's Aperture, iPhoto, RAW Camera and Core Image engineering teams: https://www.gentlemencoders.com/nitro-for-macos/index.html
So when I mistakenly copy into the wrong folder, it's obvious which image files don't belong. So I don't need to delay while I think of the best descriptive folder name. So I don't need to sort yesterdays photos into a different folder than todays photos.
Secondly it's a matter of deleting image files. I take a look with some viewer app and use the file system to delete whatever doesn't seem worth any more attention.
Thirdly whatever trial photo software is available is probably good enough to start learning.
Another terrible design in darktable is default settings. I have no problem with options, but then you need carefully choose defaults that are selected questionable here - pure exhibitionism right after opening software.
I don't agree with your statement about developing by photographers. If so, there is higher probability that they would focus on UI with more aesthetic care than coders would do.
Lack of AI masking is too expensive to use by professionals. You simply cannot afford to mask manually bunch of images. Wider adoption among photographers is simply impossible.
1. 100$ is a pocket change for many people. Depending where you live. I think it is enough of a barrier to force a thoughtful action. Also, to compare, cheapest certificates for signing windows software are like 450$/year. Microsoft has a hosted service now for $15/year, but that is still in beta. Both of those options are significantly more expensive than 99$/year Apple charges.
2. No, Apple does not deny randomly access to developer accounts. Also, this is quite besides the poin.
3. Code Signing has absolutely nothing to do with submitting your app to app store, or anything Apple has to approve of. Like, why even write about something you have no idea about?
4. True. Not sure how it is relevant.
I've literally had to argue with Apple for months to get my account approved.
It was not a fun process, eventually they did grant me an account, but then just keep up an old hobbyist game I was wasting that 100$ a month.
I ended up just recompiling for WebGL and uploading to itch.
Going through the hoops for code signing and getting account doesn't make sense if OP isn't going to sell it on the app store.
Finally, it's not like this is the only raw editor.
How many people.
A: Will pick an open source feature lite raw editor which isn't going to be as good as Darktable or Lightroom.
And
B: Aren't willing to build from source or literally run a single terminal command to run the unsigned version.
Now, if OP wants to sell it on the app store then I'm completely wrong. But otherwise it's not a good use of time.
You're free to reach out to OP and offer them 100$ + another 300$ for time spent having to apply for a Developer account.
For example, your choice of demosaicing method can make a tangible difference in finer details: some methods would make them less noisy (better for some styles), others would better preserve finer details (better for other styles). Abstracting it behind one “more detail—less detail” slider isn’t going to work because “detail” can mean a multitude of things, of which sometimes you want one and not the other, and inventing new sliders with user-friendly but inscrutable labels a la “brilliance”, “texture”, and so on, can only get you so far.
There are shades between simplicity vs. control, of course, and so I am curious to know the answer from the horse’s mouth so to speak: to what end they choose to compromise simplicity.
Almost no professional photographer will care about the intricacies of the demosaicing algorithm, or the choice between a dozen different denoising modules, and Lightroom is entirely correct in not giving you a zillion knobs to adjust things that have no effect on image quality except in the rarest of cases. In 99% of cases the controls that matter are:
* Basic exposure/shadows/contrast etc
* Curves/levels for more control if needed
* White balance
* Cropping, obviously
* Cloning/healing brush
* Simple knobs for sharpening and NR
* Level/perspective adjustment
* Lens aberration correction (most of the time no manual input needed if the lens is in the batabase)
If the white balance is set wrong in-camera, then the JPG just came out all blue. It's effectively a black and white photo (albeit in shades of blue), and there's nothing to be done about it. Shot in RAW, the photo can be made color again, extremely easily and quickly.
In fact it gets worse, not better, if you on the day try to adjust the white balance, as you go from outdoors to indoors. Not to mention if you change from flash and back. Auto is safer, but when it's wrong, the photo is unusable, and the moment is gone.
But my DSLR is now over a decade old. Maybe "auto" has gotten much better?
So yeah, for me the main thing is to be able to post facto adjust white balance, which JPG does not support. (if you've done it with both JPG and RAW, you know what I mean when I say "does not support")
The camera makes all those decisions even when shooting raw -- and there are stored in the raw file. So, by default, processing a raw file witout doing any tweaks will get you the jpeg you would have gotten.
My camera (Nikon) -- and I assume the others -- will even store both the RAW and the JPEG, so you don't even have to go through the automatic conversion step if you don't want to.
RawTherapee I uninstalled almost immediately because it crashed a few times and the UI didn't seem to jive with what I wanted to do.
Despite DarkTable's horrific interface and hostile developers I keep it around because I can often beat it into submission (but what a chore that is). And that's the thing. Even if I were shooting JPEGs DT's interface would still be a problem.
I much prefer the control darktable gives me now.
One of my complaints dealt with how unintuitive the sliders are. There's no additional control gained by making the UI widgets difficult to deal with.
Another dealt with trying to set color temperature. There are two places color temperature can be set and they'll both conflict with each other. The newer module is absurdly complex. It's great if you're writing a dissertation on color rendition but less great if you're trying to be productive.
Sure there's more control offered by having ten different demosaicing algorithms to choose from. Unfortunately I can't think of a time when I've needed or wanted that control. Maybe if I shot Fuji or Sigma. But I don't. And most folks don't.
Presets and history are a nightmare. Items in the history widget get aggregated so it's difficult/impossible to pick out individual steps. If you give labels to the actions in a preset (my terminology is off because I've not used DT much in a while)… sometimes they work. Sometimes they don't and things don't appear to pick up the label/group/whatever it's called. If memory serves I had to apply presets in one module to have them visible in the develop module.
The vestigial DAM stuff… ugh.
There's no obvious A/B split views.
Perhaps the most obnoxious thing is that DT shamelessly apes the Lightroom interface but in reality behaves almost nothing like Lightroom. There's a TON of complexity for little-if-any improvement in outcomes.
If you look at CaptureOne you can see how easy it is to edit a raw image. Most of the time it looks like the camera jpeg without having to tune anything. But then you have the options to go in depth.
Sometimes I have a photo session where everything is to my liking, just a bit of exposure and crop. Other times I shoot in night clubs with no flash and I have multiple layers of masks for a single photo.
A UI with decent defaults goes a long way into making a complex app easy to use.
DCP Tone curve
DCP Base table
DCP Look tableI will say that “auto” is pretty decent on the phones in most common lighting scenarios like sunlight/shade/outdoor/tungsten/fluorescent—white point is an entirely subjective thing that cannot be reliably determined automatically, so in my experience you rarely get the correct rendition of, say, bright pink clouds at sunset, or a book with pink pages (the phone would think it must be the some weird lighting that should be corrected for, because obviously a book can only have nearly white pages, right?), etc.—but due to physical limitations of sensor size and inferior optics the phone is worse than even a decade-old APS-C DSLR in most regards overall.
I used to work the Exposure tab curves or Lab* adjustments but that no longer seems like a good use of my attention.
See, you are saying “want to post-process”, which to me says that there is a different priority present rather than just “simplicity and ease of use”.
If the priority is “making the photos look the way you want them to look”, then we are in a territory where it is not as simple as “this tool is easy to use and therefore a better choice than that tool”.
You can want post-processing, but also don't want to spend 50 hours to learn a tool. Sometimes you just want "make it look close to the in camera jpeg, but let me adjust to exposure"
It's not that complex.
When you say “simplicity and ease of use” I think that includes taking the photos, and if I can defer some decisions that might make the overall process simpler.
This concept is pretty common, not even limited to photography.
> Another dealt with trying to set color temperature. There are two places color temperature can be set and they'll both conflict with each other.
There are at least three colour temperature / white balance controls in RT; more if you count output colour space primaries, Lab space curve, etc. as ways of creative white balance control.
I’m not sure I see any particular conflict between them. They all do different things; some of them are more relevant if you want to achieve the most precise representation (e.g., you are digitizing analog prints or paintings), others are more relevant if you are going for a creative look of your own.
It’s probably best to consult RawPedia[0], but as far as I understand:
— One of them controls raw data interpretation, and affects how different tools work down the line (e.g., highlight recovery or targeting sky with wavelets). As far as I understand, you probably want to keep this one technically correct and as close as possible to the true neutral white/grey point; at this step you are helping the tool do the rest of its job and not trying to achieve a look. If you use a colour card, a DCP profile, etc., then you know exactly what to set White Balance controls to.
— There are a couple of controls under CAM model, which you may or may not be using depending on your profile. With scene illuminant you set… well, scene illuminant (the light you have in your scene), and viewing conditions allow you to shift colours to make it look right if you know your photo will be viewed in an environment with particular light.
— Then, of course, you have dozens of different ways of creatively controlling perceived white balance via different curves or CLUT; I think these are most handy if you are going for a look.
I don't have DT under hand to check, but there are two controls which, if active at the same time, will produce a warning along the lines of "wb is already set in [the other control]".
edit: I think [0] describes the issue
[0] https://discuss.pixls.us/t/white-balance-applied-twice/34949
I tried Rawtherapee first as they actually sign their (macOS) app. Unfortunately it was too unstable to be useful.
(Later versions may have moved newer functionality to a different tab or menu item, is that what you mean?)
I have never seen Lightroom (or C1, for that matter) as compelling at all ever since I started using RawTherapee. Unlike, say, InDesign, which is legitimately a difficult to replace professional tool with incredible capabilities, Adobe’s raw image processing offering looks incredibly dumbed down.
On linux, the default Gnome image viewer is nice but you can't make adjustement and when deleting a file, the file is not remove from the NAS directory (need a manual refresh). With Gthumb it works for deleting files but the crop tool and the overall app is not as nice. Anyway I'll continue to look for my perfect app or for the default Gnome viewer to update its features (I think it is in active development)
my advice would be to look for that and not a raw image processing tool
that incidentally happens to also handle cataloguing (inevitably in a
half-baked way).
About the worst thing I can say about Lightroom is that it didn't reliably work with my iPhone. Otherwise it did everything I needed in terms of tagging, presets, and organizing the pictures on the file system.
Meanwhile darktable creates freaking sidecars for every picture while it relies on an SQLite database for tracking history just like Lightroom does.
I have never seen Lightroom (or C1, for that matter) as compelling at all ever since
I started using RawTherapee.