OP isn't just making a straw man argument, they're making the entire straw family.
> Those defending the poor performance of vibe coded apps have parroted the famous lines AI guys used in the past - “It is only days old” and “this is the worst it gets”. They make the mistake of equating human app development time with generative AI development time (hundreds of times faster). A vibe coded app is, in human hours, not only a few days old but closer to two or three years old. If a brush engine or transform tool works badly after the human equivalent of three years of development that is a demonstration of bad planning or an AI model not doing a good job.
I legitimately cannot parse the logic behind this argument.
It’s really badly put but I thiiiink they’re trying to say that due to the pace you can work with AI, you can have something technically complex but badly engineered quickly. Like three years of bad engineering in an afternoon? Or something? It’s still bad as even three months of bad engineering would normally contain far more foresight than a vibe coded weekend nonsense project
I see it like this: because greenfield coding is strong with AI, but improving/large rewriting an existing codebase is getting progressively harder and worse, if you don't get the architecture right in the first try, improve them in later rounds will be very hard,if not futile. Thus the poor performance is not a typical day 1 problem as with human coding, but a pestilence.
That has not been my experience. Agents love to refactor. They also love to ridiculously overengineer it and run the new version as shadow in the background for several months before adoption.
The article is strangely supportive of Adobe, far beyond a normal person's sense of ridicule.
Software development is about revenue, but specifically Adobe's revenue:
> Automated or semi-automated cloning to steal revenue away from developers
Pay us, don't innovate:
> They’re spending minimum $2K to vibe code a Photoshop competitor when Affinity Photo is free and Photoshop costs $200 a year. Not very smart economics.
> Commercial apps innovate and lead. Open source apps like Blender follow the lead. Innovation costs a lot of money.
> Instead of hurling hate speech and bile at thousands of software engineers who lead innovation at large companies (Adobe, Autodesk, Maxxon, Celsys, and others), show them some gratitude for their efforts. They are the engineers who innovate and everyone else follows. Just because they work at a big company doesn't mean they are committing human rights abuses. We have a generation brainwashed by envy and hate instead of urging everyone to be greater and bigger.
Why do people hate Adobe more than other companies? What's the difference? Only sensitive insiders would pretend to ignore decades of enshittification.
> Clip Paint Studio, Maya, 3ds Max and Cinema 4D, btw, are all subscription applications. But because those apps aren’t in demand by pirates Autodesk has not experienced the hate and bullying campaigns that Adobe has received.
People noticed the "underage children", but defending Adobe customer support is even more ridiculous:
> And when it comes to customer support things become even sadder looking. One or two guys in a Discord chatroom cannot support a global user base, especially when those two guys are hunting for underage children to speak to on Discord (yes, I have seen what vibe coders do).
> "The performance of their brushes collapses once the canvas size is over 8K x 8K pixels. Photoshop’s brushes work smoothly even when you’re working with 100 megapixel images out of a Hasselblad or Phase One camera"
I knocked up some quick code for drawing on an image a while back as part of a work project, but I've never actually checked how well it works on gigantic images. It calculates the bounding box of your stroke every frame, and uses an SDF-based system to calculate the pixels of the stroke (so if you draw quickly, it will correctly join up the points). I didn't focus too heavily on performance, other than making sure it was good enough to not slow down the app, but since its speed only depends on how much you're actually drawing per frame, rather than the size of the canvas, I'd imagine it would hold up reasonably on gigantic images?
Some valid points but I'm not sure if I agree with the brush engine point. I don't think it's the most important feature of Photoshop to many, so not making that one right first is not that surprising.
Making a good brush engine is indeed a very hard problem, especially when you have undo history, high res canvases, layers, zoom etc. I just don't see it of such importance
My team are and the full Adobe plans and for our use cases it's also not important. We do have a few workflows that dive deeper into niche tools only Photoshop has though, but I can see competitors making streamlined apps for specific use cases that can compete. It's actually kind of tempting to see if I can get something done over a couple nweekends and see if it'd be usable
I've been looking at the posting history of the creator and I think we differ in terms of philosophical perspective. His reddit account was 'ai_art_is_art' which feels shock jokey-ish even if you agree with that, and I also don't think there's legal longevity in copying Adobe's products 1:1. I might put some development time into another open source editor
I hear that groups are using AI to reverse-engineer video games, targeting PlayStation, Xbox and Nintendo, with some success, and they’re now able to run natively on Linux and Windows.
Add that to the fact all 3 companies have seen massive declines in sales, and hardware costs increase quite dramatically, this may be the end of the line for the console wars.
Interestingly, Sony released some of their exclusives on PC a few years ago and they sold extremely well. So what did Sony do? Announce that no more exclusives will port to PC. Nice work, guys!
No one hates Adobe more than their paying customers. I hope everyone involved in the move to subscriptions, including the gross cancellation fees, know that they killed the company.
The backlash against vibecoded decomps and using a lot of compute for unimpactful projects was met with the counterargument "why can't you vibecode something useful like Photoshop instead?"
No, the backlash was because vibecoded decomps were trash, incomplete, and quickly abandoned once the attention rush wore off. You'll never guess what's happening to these!
To answer the question regarding bloat in adobe software just look at the creative cloud app and how you have to use it to update and install individual applications. Look at the Adobe Genuine Software service that runs on Windows. And that’s just the tip of the iceberg.
The tone is angry but understandably so. The labour, creativity and innovation of millions is being stolen and then belittled by grifters shilling slop. Art, design, math, music, software, there's nothing they won't claim they can now copyright-launder and somehow magically render obsolete. But look at any of it and it's never more than the simulacrum of real work. It always looks like the real thing, particularly to the untrained eye but it's always a tissue-paper of stolen work and half-assed, half-working, sloppy output. The prompters are lazy and so are the models.
These are powerful, interesting tools but the money, the hype and the grift is just intollerable, a pox on it all. If you're a professional you cannot ignore or avoid it, but you can push towards a future where the commodification of it - open models, open weights, open hardware - turns it into a low-margin overlay on real intellectual work and the hype/grift machine can move on to whatever intellectually bankrupt boondoggle they find next.
Guess, that cruising phase of a ton of careers is cut short. Its back to the trenches, to work and shovel till death comes for us keyboard slaves. If some singularity demon can undo your moat, you never had one to begin with.
The false claims of the projects capabilities along with the use of alts to astroturf and spread false narratives is really telling about the intentions and mindset of the authors.
Whether or not AI ever gets good enough to polish their turds, it's worth considering their motives.
When in doubt accuse anyone hurting your corporate profits of being a terrorist and/or paedophile. It's the Epstein class's favourite tactic, but every accusation from them is a confession.
If I was Adobe's marketing department, I wouldn't be sure whether to run around quietly seeding such stories, or be completely silent and hope people won't notice the moat is getting thinner.
> And when it comes to customer support things become even sadder looking. One or two guys in a Discord chatroom cannot support a global user base, especially when those two guys are hunting for underage children to speak to on Discord (yes, I have seen what vibe coders do).
> Instead of hurling hate speech and bile at thousands of software engineers who lead innovation at large companies (Adobe, Autodesk, Maxxon, Celsys, and others), show them some gratitude for their efforts. They are the engineers who innovate and everyone else follows. Just because they work at a big company doesn't mean they are committing human rights abuses. We have a generation brainwashed by envy and hate instead of urging everyone to be greater and bigger.
I mean okay vibe coded photoshop clones is one thing but arguing your enemies are pedos and you're good guys that deserve all the gratitude in the world is not a great way to make an argument.
There are many good valid points in the whole insanity of the whole trend imo but this article drifts a LOT.
Here are 2 quotes, which give you a pretty good idea about the whole article:
> They’re spending minimum $2K to vibe code a Photoshop competitor when Affinity Photo is free and Photoshop costs $200 a year. Not very smart economics.
and
> Instead of hurling hate speech and bile at thousands of software engineers who lead innovation at large companies (Adobe, Autodesk, Maxxon, Celsys, and others), show them some gratitude for their efforts. They are the engineers who innovate and everyone else follows.
Heh - good choices. Seems like a poorly reasoned article then.
Accepting the premise of 2000$ spent (I doubt it), the author completely forgets that the collectively saved amount far exceeds the 200$ of a single subscription.
Anyway, if I worked at Adobe, or any company with roughly the same business model, I definitely would feel nervous about now. And I still feel sympathy for the devs. But let’s be honest, Adobe principally benefits its shareholders and other dead weight, not so much the developers.
There’s no way you or anyone else actually thinks that first point is rational right?
That price comparison makes no sense. Users aren’t each paying $2,000 to use the alternative. It’s already paid and now free to use. People can chip in some money and time to contribute and build off the existing work to improve it.
How could anyone read that and think that’s a good point?
I suspect most of these projects are started by people who get Claude Code at work, and just want to make sure that they spend all of their tokens before the weekly reset. In that case, cost is not relevant.
The project the blog post is attacking is Artcraft, spearheaded by HN's most vocal AI evangelist zealot. He's stated that he's financing the thing with the revenue from his $3, no $4, no $5M/year AI image generator.
Which is why the salient part of the post is "Unless your intention is just to generate headlines and promote an AI business." Because that's exactly what this is. A cynical scheme to draw attention to his commercial services.
i can imagine reading something similar about Linux 30 years ago.
> oh nooooo, what's going to happen to all the Unix vendors with the systems they sell for thousands of dollars?
adobe has fucked their customers for over a decade, is time they're fucked back. I won't miss them. Hope their employees can find jobs at better companies quickly.
also:
>They’re spending minimum $2K to vibe code a Photoshop competitor when Affinity Photo is free and Photoshop costs $200 a year. Not very smart economics.
very smart economics, because they do it once and get to have it forever, for free, for everyone. it's charity.
autoresearch can solve a lot of these issues and i am sure people are usually heavily heavily subsidized subscriptions
article seems very salty and im not sure why
also the economics is you open source something and its saves every one money and opens up more cross compile support, i bet this will scare companies to support linux more properly before linux users just start getting very good at cloning
This. I had to create a temporary card with my banking app (Monzo), then I could switch out the payment method on my adobe account. Then I could cancel the temp card and then finally Adobe can die whilst trying to charge my cancelled card.
They wanted me to pay £80 to cancel, I’d had the account for 10 years so no idea what the cancellation related to.
For anyone in this situation - first upgrade your subscription to a different tier, wait a day for the account to reflect you are on a new test, then you cancel with the 7 day right to cancel. That way you avoid early cancellation fees
I've told this story before but I had to tell their scum support that I had cancer before they'd cancel my account. They just flat out refused. It was almost impressive in a way.
I never understood that problem. Why can't you cancel the credit card used for the subscription? (Yes, it's a bit of a pain to move other subscriptions to another card, but also a great opportunity to clear out some other leeches.)
In a couple weeks this ‘pirated’ software (not sure what’s pirated about it btw) will have good brush performance and will be handling 100 megapixel images easily precisely because of this post.
I for one am not grieving the saas rent seeking model, but as most folks here (I assume) my livelihood depends on it since that’s how my employer finances my salary. So yes, I’m concerned, but also am aware that technology cannibalizing technology is a thing and we have to move on and make ourselves useful in this new world.
Yeah, exactly what i thought. Behind all the rant and hate in the post (that i suppose may be driven by fear or the shock that one's greatest achievements are now close to a commodity), there is genuinely valuable criticism that an agent can be prompted to address.
The bottleneck is no longer the AI's ability to complete tasks, improve performance or research and develop new approaches, but in verifying what they built (at least in heavily vibe-coded projects), so beyond verifier loops such feedback is what leads those kinds of software to improve.
I mean this crap will die hopefully die off when they get priced out of the needed tokens and their customers realise that there’s a limit to how many times you can write “Claude fix this bug plz” before the “devs” actually need to wade into some real, and at the scale of a vibed up PS clone, extremely confusing engineering.
So many people in that thread saying "No one is going to vibe code a Photoshop replacement", and that was only 5 months ago. What the hell is going to be happening in 10 years?
It's not that you can build a photoshop clone by spending 2000$ in claude API credits. Adobe is going to do a better job than you.
What works out easier is that you can build a version that only has the features you need, with a ~20$ subscription of any of the good AI coding harnesses.
The basic photo editing tools were set in stone decades ago. Adobe has spent years tweaking things 1% of their userbase wants like 3D manipulation. Once somebody makes a basic clone which does the main basic tools well then its game over for Photoshop.
I wish major software like office and adobe (photoshop) didn't come with mandatory subscriptions, cloud integration, and AI. The most ironic outcome would be a good-enough AI-generated photoshop clone that doesn't have user-facing AI integration.
At the moment they seem to be cloning the UI and functionality. That is serious enough of a threat but it has its limits, particularly with old file formats like PSD or old MS Office that were a direct 1:1 mapping of C memory structs onto the disk.
But that won't be the case for too long I think. AI is nowadays capable of doing decompilation based on prior knowledge and educated guesswork - it isn't much of a stretch to assume that runtime code decryption / obfuscation will fall too. When humans can do it (see e.g. extremely talented people like the infamous Empress), AI will be able to as well.
> it has its limits[] with old file formats or old MS Office
Those problems map extremely well to agent loops. Take a cheap model (like Luna, which can basically run all the time on a Codex subscription, or V4.1 Flash), give it a goal like <1% pixel difference between it's own implementation and what the oracle produces, and it can run for hours or days probing the oracle, verifying it's progress and handling edge case after edge cast until it's effectively perfect.
Key idea is to remove the human from the loop.
Don't let the agent build something, tell you what it did and you point out all the issues. It does not scale, leads to fatigue and impatience and limited uptime for the agent. Build clear verifier loops the agent can rely on, let it run and such edge cases are well within reach.
That approach has its dangers as well. You might end up with a situation some researchers many years ago ended up [1], the ML algorithm gave something that fulfilled the task but only due to random side effects.
That is why, for now, you need a human in the loop that can check if the AI is following a nonsense tangent.
141 comments
[ 0.26 ms ] story [ 24.3 ms ] thread> Those defending the poor performance of vibe coded apps have parroted the famous lines AI guys used in the past - “It is only days old” and “this is the worst it gets”. They make the mistake of equating human app development time with generative AI development time (hundreds of times faster). A vibe coded app is, in human hours, not only a few days old but closer to two or three years old. If a brush engine or transform tool works badly after the human equivalent of three years of development that is a demonstration of bad planning or an AI model not doing a good job.
I legitimately cannot parse the logic behind this argument.
Software development is about revenue, but specifically Adobe's revenue:
Pay us, don't innovate: Why do people hate Adobe more than other companies? What's the difference? Only sensitive insiders would pretend to ignore decades of enshittification. People noticed the "underage children", but defending Adobe customer support is even more ridiculous:I knocked up some quick code for drawing on an image a while back as part of a work project, but I've never actually checked how well it works on gigantic images. It calculates the bounding box of your stroke every frame, and uses an SDF-based system to calculate the pixels of the stroke (so if you draw quickly, it will correctly join up the points). I didn't focus too heavily on performance, other than making sure it was good enough to not slow down the app, but since its speed only depends on how much you're actually drawing per frame, rather than the size of the canvas, I'd imagine it would hold up reasonably on gigantic images?
Making a good brush engine is indeed a very hard problem, especially when you have undo history, high res canvases, layers, zoom etc. I just don't see it of such importance
I'd consider adding/improving the features directly there, unless a more streamlined alternative is the goal.
I wonder if the project only takes contributions that aim for parity with the Adobe software or also new features and improvements.
Add that to the fact all 3 companies have seen massive declines in sales, and hardware costs increase quite dramatically, this may be the end of the line for the console wars.
Interestingly, Sony released some of their exclusives on PC a few years ago and they sold extremely well. So what did Sony do? Announce that no more exclusives will port to PC. Nice work, guys!
No one hates Adobe more than their paying customers. I hope everyone involved in the move to subscriptions, including the gross cancellation fees, know that they killed the company.
When they did vibecode Photoshop, well...
What are you basing that on? That doesn’t seem to be true at all
These are powerful, interesting tools but the money, the hype and the grift is just intollerable, a pox on it all. If you're a professional you cannot ignore or avoid it, but you can push towards a future where the commodification of it - open models, open weights, open hardware - turns it into a low-margin overlay on real intellectual work and the hype/grift machine can move on to whatever intellectually bankrupt boondoggle they find next.
Whether or not AI ever gets good enough to polish their turds, it's worth considering their motives.
Felt like that undermined some of the more cogent points the article was making.
only a serious business can replace. figma, canva and other tools. but these are new tools in the same industry.
> Instead of hurling hate speech and bile at thousands of software engineers who lead innovation at large companies (Adobe, Autodesk, Maxxon, Celsys, and others), show them some gratitude for their efforts. They are the engineers who innovate and everyone else follows. Just because they work at a big company doesn't mean they are committing human rights abuses. We have a generation brainwashed by envy and hate instead of urging everyone to be greater and bigger.
I mean okay vibe coded photoshop clones is one thing but arguing your enemies are pedos and you're good guys that deserve all the gratitude in the world is not a great way to make an argument.
There are many good valid points in the whole insanity of the whole trend imo but this article drifts a LOT.
Here are 2 quotes, which give you a pretty good idea about the whole article:
> They’re spending minimum $2K to vibe code a Photoshop competitor when Affinity Photo is free and Photoshop costs $200 a year. Not very smart economics.
and
> Instead of hurling hate speech and bile at thousands of software engineers who lead innovation at large companies (Adobe, Autodesk, Maxxon, Celsys, and others), show them some gratitude for their efforts. They are the engineers who innovate and everyone else follows.
Accepting the premise of 2000$ spent (I doubt it), the author completely forgets that the collectively saved amount far exceeds the 200$ of a single subscription.
Anyway, if I worked at Adobe, or any company with roughly the same business model, I definitely would feel nervous about now. And I still feel sympathy for the devs. But let’s be honest, Adobe principally benefits its shareholders and other dead weight, not so much the developers.
That price comparison makes no sense. Users aren’t each paying $2,000 to use the alternative. It’s already paid and now free to use. People can chip in some money and time to contribute and build off the existing work to improve it.
How could anyone read that and think that’s a good point?
Which is why the salient part of the post is "Unless your intention is just to generate headlines and promote an AI business." Because that's exactly what this is. A cynical scheme to draw attention to his commercial services.
> oh nooooo, what's going to happen to all the Unix vendors with the systems they sell for thousands of dollars?
adobe has fucked their customers for over a decade, is time they're fucked back. I won't miss them. Hope their employees can find jobs at better companies quickly.
also:
>They’re spending minimum $2K to vibe code a Photoshop competitor when Affinity Photo is free and Photoshop costs $200 a year. Not very smart economics.
very smart economics, because they do it once and get to have it forever, for free, for everyone. it's charity.
article seems very salty and im not sure why
also the economics is you open source something and its saves every one money and opens up more cross compile support, i bet this will scare companies to support linux more properly before linux users just start getting very good at cloning
In a couple weeks this ‘pirated’ software (not sure what’s pirated about it btw) will have good brush performance and will be handling 100 megapixel images easily precisely because of this post.
I for one am not grieving the saas rent seeking model, but as most folks here (I assume) my livelihood depends on it since that’s how my employer finances my salary. So yes, I’m concerned, but also am aware that technology cannibalizing technology is a thing and we have to move on and make ourselves useful in this new world.
LLMs remain a microwave.
https://news.ycombinator.com/item?id=48177228 - (Where Are the Vibecoded Photoshops?)
It's not that you can build a photoshop clone by spending 2000$ in claude API credits. Adobe is going to do a better job than you.
What works out easier is that you can build a version that only has the features you need, with a ~20$ subscription of any of the good AI coding harnesses.
But that won't be the case for too long I think. AI is nowadays capable of doing decompilation based on prior knowledge and educated guesswork - it isn't much of a stretch to assume that runtime code decryption / obfuscation will fall too. When humans can do it (see e.g. extremely talented people like the infamous Empress), AI will be able to as well.
Those problems map extremely well to agent loops. Take a cheap model (like Luna, which can basically run all the time on a Codex subscription, or V4.1 Flash), give it a goal like <1% pixel difference between it's own implementation and what the oracle produces, and it can run for hours or days probing the oracle, verifying it's progress and handling edge case after edge cast until it's effectively perfect.
Key idea is to remove the human from the loop. Don't let the agent build something, tell you what it did and you point out all the issues. It does not scale, leads to fatigue and impatience and limited uptime for the agent. Build clear verifier loops the agent can rely on, let it run and such edge cases are well within reach.
That is why, for now, you need a human in the loop that can check if the AI is following a nonsense tangent.
[1] https://www.damninteresting.com/on-the-origin-of-circuits/