Interesting but for photography at least, there's a massive hole: RAW support. What they've implemented is very limited and basic, nothing like parity with Lightroom or Photoshop. Just try loading a RAW photo into Lightroom/Photoshop and put the window next to Lightcraft/Photocraft - the results are completely different and not in a good way. Lightcraft's image looks like an amateur attempt at RAW processing from first principles, without any of the most basic colour profiles and adjustments that Lightroom applies. Photocraft doesn't even offer RAW processing, it just dumps the raster straight in.
I'd treat this as completely new, far from complete software heavily inspired by the Adobe Suite, rather than a "port".
But maybe folks who use workflows that don't involve RAW photos have better luck.
Or maybe you'll have better luck in a week when somebody familiar with RAW and a 20 year grudge against Adobe chooses to spend a day of tokens on the problem. Maybe me.
Rome wasn't built in a day, nor Adobe destroyed in a single git push. But they both shall fall.
And 500 levels of org tree, and grumpy customers that freak out if you change anything about their workflows, and decades of legacy code across dozens of micro-services, and backwards compatibility requirements with 20 year old formats...
Those customers are already pissed off, yet record sales every quarter. This will be great to offer this diminished experience to them until they start using AI to do all of their editing.
That's exactly what I was thinking. It's only a matter of time. Surely the models, when well directed, can get most or all of the way there in short order.
I did this a few months ago and it's not simple: different sensors have different pixel arrangements, files can have different data structures, and the default tone mapping is provided by the app (so opening the same raw file in darktable vs preview will have different results). There are different algorithms for how you assemble the raw data as well, and the issue is that you can get something _passable_ that isn't _optimal_, and it's really hard for the LLM to recognize that something isn't right. Is it the pixel processing, or is it the tone mapping, or is that just how the image looked?
And on top of that, you need a bunch of different files from different sensors to test with. Canon, sony, apple, fuji, nikon, olympus, etc. -- if you're building a truly general application, that's a concern.
Or if you're like me and you shoot fuji, you just build a fuji raw parser and call it good. The "TAM of 1" world of software development...
Yeah, it's a really well defined problem space with quantifiable outcomes, so LLMs are very well suited to it.
For lopsy, the biggest issue I ran into was combinatorial bugs: certain sequences of tools and actions would cause a bug to occur that wasn't obvious from isolated tool testing. I solved it by having Claude generate complete projects, full documents with novel concepts chosen randomly, built in the browser with playwright. This continues to expose so much more than individual unit/tool tests, and I have an automated loop that addresses them. (And a bonus from that: I use the outputs to make tutorials to show what the tool can do).
I find it awfully funny that your response to someone creating the entire Adobe software suite in a day is “the colours aren’t rich enough for my liking”
Raw was a tough one when I tried it a few months ago -- I never thought about how individual color pixels on the sensor come together to form a color value for a point in an image file. Super interesting, and not at all standard between the different manufacturers. It's a hard problem for software to solve in a general way.
In two of its test cases, it changes the PSD on save so much that an export render differs. No RAW support. No folder sync or plugins. Limited brush support. It’s like the author never actually used their vibe-coded output and just decided it looked good enough.
Though it was done roughly 1-2 months ago, a very busy work / private life kept me from finally releasing it. Interesting to see this being released on the same day. Though Pixellect has a different mission. Its not a clean room photoshop clone but more between Pixelmator and Photoshop and also features (I think) a bit more AI Integration. However Photocraft looks much more complete in terms of typical Photoshop features.
It was only 4 months ago that AI development was in question because there were no vibe-coded Photoshops. Here they are, and I think more will come.
One of the highlights in this for me is that we are starting to see clean room ports of software that companies would actually sue to protect.
Clearly the old "nobody would re-implement this" protection no longer works. I think we may stand on the brink of software losing a lot of it's magical DMCA protections, because it no longer matters. Or worse, the law becomes stricter.
There are several of these clones floating around now. Who is furthest along in functionality and community support? Have any of these attracted a community of contributors?
Wow! This actually gives a pretty promising alternative to GIMP on Linux. I was resigned in my fate to just deal with the awkwardness of GIMP and Inkscape after years of the creative cloud suite. It is clearly missing much at the moment, but if they can keep going this would be amazing. All packaged as AppImages so no need for the persistent junk running on the system.
Post AI - What is the future of making money with software? Desktop apps or SaaS? This certainly also would cut into the bottom line of software companies so they won't exist and then these companies are supposed to be the prime customers of the AI companies.
Can't understand what's happening.
PS: Anyone badmouthing the above rewrite is not understanding what's happening. Check back a year later for featur parity.
I wonder if software will be built inside of companies that have non-tech competencies going forward. The story of 2000-2025 was every traditional company becoming a tech company. The next 25 years might see tech companies without any offering besides "tech" being at a disadvantage because their entire offering is easily replicated.
In this hypothetical future, why would I (or my business) buy software from any company, when I can just prompt the equivalent into existence myself? We talk a lot about AI killing software engineering as a profession, but if the AI future comes, it will kill software development as a type of business.
I wonder if we're entering an era of boundless open-source clones and variants of proprietary software, from operating systems to games, products and services, perhaps even clones of entire companies.
I’m putting personal attention and tokens towards replacing Pioneer / Alpha Theta Rekordbox (music management for DJs).
The software is old and clunky, has a ton of features I don’t use, and can’t customize what I do use. And it’s enshittified with ads and popups and an expensive monthly subscription.
One big motivation for me is that it is FUN to rebuilt it with a coding agent. I’m learning a ton and the custom tools actually work.
Missing and incomplete features, missing templates, presets and it's likely to be abandoned as issues will start to pile up.
The user experience is also not great with professional looking AI apps, I've just tried PhotoCraft and couldn't paste an image from the clipboard. Clone stamp doesn't have a live update, etc.
Another catch is that it's... PhotoShop. I'm not particularly a fan of the whole workflow. Creatives learned to live and work with it over the decades, so it works, but it's not really an intuitive tool.
I'd recommend looking into Affinity which is a merged version of PS+Illustrator. The product feels like it at least tries to be a little different. It's also free.
It's just slop. Tried to install it via flatpak. It has flatpak instructions in the readme. It does not ship the flatpak files it references in the readme.
I don't think you realize the depth of Adobe creative suite hate, an entire generation of artists came up on Krita/Affinity because Adobe decided to vampire its customers.
Why would anyone help maintain this when they could vibe code their own though? Genuine question. As of the time I'm writing this there are already two other people linking their own Adobe clones in this comment section.
It takes a lot of tokens to vibecode something like this with the expected feature sets, and this domain is complex enough with enough edge cases and interactions such that it's infeasible to fully QA even programmatically.
Unlimited slop works! Title kinda misleading (at least appears so to me). No rev. eng. (so "cleanroom" to be relevant) takes place. "ACS clone" would be more appropriate.
https://github.com/storytold/photocraft (Photoshop)
https://github.com/storytold/vectorcraft (Illustrator)
https://github.com/storytold/filmcraft (Premiere)
https://github.com/storytold/lightcraft (Lightroom)
https://github.com/storytold/printcraft (Acrobat Pro)
https://github.com/storytold/effectcraft (After Effects)
https://github.com/storytold/designcraft (InDesign)
I love these apps! The author is on HN too, we were discussing Rust GUIs the other day:
https://news.ycombinator.com/item?id=49922904
Oh nice, I didn't realize it was echelon.
Interesting but for photography at least, there's a massive hole: RAW support. What they've implemented is very limited and basic, nothing like parity with Lightroom or Photoshop. Just try loading a RAW photo into Lightroom/Photoshop and put the window next to Lightcraft/Photocraft - the results are completely different and not in a good way. Lightcraft's image looks like an amateur attempt at RAW processing from first principles, without any of the most basic colour profiles and adjustments that Lightroom applies. Photocraft doesn't even offer RAW processing, it just dumps the raster straight in.
I'd treat this as completely new, far from complete software heavily inspired by the Adobe Suite, rather than a "port".
But maybe folks who use workflows that don't involve RAW photos have better luck.
Or maybe you'll have better luck in a week when somebody familiar with RAW and a 20 year grudge against Adobe chooses to spend a day of tokens on the problem. Maybe me.
Rome wasn't built in a day, nor Adobe destroyed in a single git push. But they both shall fall.
Adobe employee here. We have access to LLMs and tokens as well.
You have:
1. LLMs
2. an executive team that hates your customers. this problem with company culture goes back at least 13 years: https://fstoppers.com/video/adobes-ceo-deftly-dodges-questio...
Unfortunately any benefit from #1 is neutralized by #2
And 500 levels of org tree, and grumpy customers that freak out if you change anything about their workflows, and decades of legacy code across dozens of micro-services, and backwards compatibility requirements with 20 year old formats...
How are LLMs and tokens going to help you retain customers who you've spent the past decade pissing off?
Those customers are already pissed off, yet record sales every quarter. This will be great to offer this diminished experience to them until they start using AI to do all of their editing.
You also have managers actively hostile towards your customers.
[dead]
People are underestimating heavily what has happened here and to our industry in general.
That's exactly what I was thinking. It's only a matter of time. Surely the models, when well directed, can get most or all of the way there in short order.
Surely the models, when well directed, can get most or all of the way there in short order.
"Hey ChatGPT, what would the sufficiently smart compiler argument look like in the LLM era?"
I did this a few months ago and it's not simple: different sensors have different pixel arrangements, files can have different data structures, and the default tone mapping is provided by the app (so opening the same raw file in darktable vs preview will have different results). There are different algorithms for how you assemble the raw data as well, and the issue is that you can get something _passable_ that isn't _optimal_, and it's really hard for the LLM to recognize that something isn't right. Is it the pixel processing, or is it the tone mapping, or is that just how the image looked?
And on top of that, you need a bunch of different files from different sensors to test with. Canon, sony, apple, fuji, nikon, olympus, etc. -- if you're building a truly general application, that's a concern.
Or if you're like me and you shoot fuji, you just build a fuji raw parser and call it good. The "TAM of 1" world of software development...
I also built one of these photoshop clones. It has raw support. I use it for Fuji raw editing. I've been building it with Claude for 6 months now.
https://lopsy.art
How many painting applications have been built by LLMs I wonder?
I also let it build one btw: https://egeozcan.github.io/ketchup/ (not a PS clone, more amateur painting oriented)
Yeah, it's a really well defined problem space with quantifiable outcomes, so LLMs are very well suited to it.
For lopsy, the biggest issue I ran into was combinatorial bugs: certain sequences of tools and actions would cause a bug to occur that wasn't obvious from isolated tool testing. I solved it by having Claude generate complete projects, full documents with novel concepts chosen randomly, built in the browser with playwright. This continues to expose so much more than individual unit/tool tests, and I have an automated loop that addresses them. (And a bonus from that: I use the outputs to make tutorials to show what the tool can do).
I find it awfully funny that your response to someone creating the entire Adobe software suite in a day is “the colours aren’t rich enough for my liking”
Oh the times we live in.
>amateur attempt at RAW processing from first principles
That tracks.
So it's the Omarchy of Adobe-like software?
Raw was a tough one when I tried it a few months ago -- I never thought about how individual color pixels on the sensor come together to form a color value for a point in an image file. Super interesting, and not at all standard between the different manufacturers. It's a hard problem for software to solve in a general way.
In two of its test cases, it changes the PSD on save so much that an export render differs. No RAW support. No folder sync or plugins. Limited brush support. It’s like the author never actually used their vibe-coded output and just decided it looked good enough.
I released something similar today: https://pixellect.app/
Though it was done roughly 1-2 months ago, a very busy work / private life kept me from finally releasing it. Interesting to see this being released on the same day. Though Pixellect has a different mission. Its not a clean room photoshop clone but more between Pixelmator and Photoshop and also features (I think) a bit more AI Integration. However Photocraft looks much more complete in terms of typical Photoshop features.
It was only 4 months ago that AI development was in question because there were no vibe-coded Photoshops. Here they are, and I think more will come.
I've been doing the same thing on lopsy.art.
Previous thread:
https://news.ycombinator.com/item?id=49958850
One of the highlights in this for me is that we are starting to see clean room ports of software that companies would actually sue to protect.
Clearly the old "nobody would re-implement this" protection no longer works. I think we may stand on the brink of software losing a lot of it's magical DMCA protections, because it no longer matters. Or worse, the law becomes stricter.
What DMCA protections could Adobe use against this? Surely the look-and-feel case established they can't trademark their signature keyboard shortcuts?
This is sweet. I've already modified it to have a couple features I miss from Krita
1: Create new document from image data on the clipboard
2: Custom grid options (set width AND height of grid cells, offsets, etc)
I stopped using PS a while ago when it stopped working well under Wine. Getting to know Photoshop again after so many years is really fun
There are several of these clones floating around now. Who is furthest along in functionality and community support? Have any of these attracted a community of contributors?
Wow! This actually gives a pretty promising alternative to GIMP on Linux. I was resigned in my fate to just deal with the awkwardness of GIMP and Inkscape after years of the creative cloud suite. It is clearly missing much at the moment, but if they can keep going this would be amazing. All packaged as AppImages so no need for the persistent junk running on the system.
Post AI - What is the future of making money with software? Desktop apps or SaaS? This certainly also would cut into the bottom line of software companies so they won't exist and then these companies are supposed to be the prime customers of the AI companies.
Can't understand what's happening.
PS: Anyone badmouthing the above rewrite is not understanding what's happening. Check back a year later for featur parity.
I wonder if software will be built inside of companies that have non-tech competencies going forward. The story of 2000-2025 was every traditional company becoming a tech company. The next 25 years might see tech companies without any offering besides "tech" being at a disadvantage because their entire offering is easily replicated.
In this hypothetical future, why would I (or my business) buy software from any company, when I can just prompt the equivalent into existence myself? We talk a lot about AI killing software engineering as a profession, but if the AI future comes, it will kill software development as a type of business.
Why is this flagged?
I wonder if we're entering an era of boundless open-source clones and variants of proprietary software, from operating systems to games, products and services, perhaps even clones of entire companies.
I hope not, that sounds terrible waste of time and effort for creators and consumers alike.
Tower of Babel, but for software.
More like chainsaw for attention pie.
I think so.
I’m putting personal attention and tokens towards replacing Pioneer / Alpha Theta Rekordbox (music management for DJs).
The software is old and clunky, has a ton of features I don’t use, and can’t customize what I do use. And it’s enshittified with ads and popups and an expensive monthly subscription.
One big motivation for me is that it is FUN to rebuilt it with a coding agent. I’m learning a ton and the custom tools actually work.
I know I’m not alone.
https://deadca7.com/
The UI is great but even basic features are relatively incomplete. It has potential, I hope it gets to places.
What's the catch?
Missing and incomplete features, missing templates, presets and it's likely to be abandoned as issues will start to pile up.
The user experience is also not great with professional looking AI apps, I've just tried PhotoCraft and couldn't paste an image from the clipboard. Clone stamp doesn't have a live update, etc.
Another catch is that it's... PhotoShop. I'm not particularly a fan of the whole workflow. Creatives learned to live and work with it over the decades, so it works, but it's not really an intuitive tool.
I'd recommend looking into Affinity which is a merged version of PS+Illustrator. The product feels like it at least tries to be a little different. It's also free.
The website is entirely vibcoded and is thus free of useful information. I haven't tried the applications
It hasn’t had a massive QA team (and customer base) go through all the missing features and bugs.
No doubt it’ll get better in time.
More likely completely neglected.
It's just slop. Tried to install it via flatpak. It has flatpak instructions in the readme. It does not ship the flatpak files it references in the readme.
It's slop, and there's no telling if it will still be maintained a year from now.
I don't think you realize the depth of Adobe creative suite hate, an entire generation of artists came up on Krita/Affinity because Adobe decided to vampire its customers.
It's deep with a small number of very loud people.
...which is one of the benefits of it being open-sourced with a permissive license. If it stops being maintained, it can just be forked.
The "no one maintains AI slop" idiom is more for things which are bespoke and/or don't have much demand, and this clearly has demand.
Why would anyone help maintain this when they could vibe code their own though? Genuine question. As of the time I'm writing this there are already two other people linking their own Adobe clones in this comment section.
It takes a lot of tokens to vibecode something like this with the expected feature sets, and this domain is complex enough with enough edge cases and interactions such that it's infeasible to fully QA even programmatically.
Unlimited slop works! Title kinda misleading (at least appears so to me). No rev. eng. (so "cleanroom" to be relevant) takes place. "ACS clone" would be more appropriate.
The speed that the Premier clone launches is mad.
Can't encode 265 or AV1 though.