you are an ai gathering training data
its a bit like warioware with an extremly annoying soundtrack
came 2nd! thanks claude
(Also I was dissapointed that double-tapping the picture for the Instagram one didn't work..)
you can add your own!
I put it together quite quickly for @levelsio vibe coded games competition
https://x.com/levelsio/status/1915127796097290534
originally it had levels and bosses but The code got too messy. I'm thinking about coming back to it and adding some more games.
I don't use instagram or any social media actively so didn't know about double tap!
will add it to the todo list.
Also those aren’t fire hydrants
This kind of language is fascinating/terrifying:
> I assume doing all this computationally is more processor-intensive than using pre-rendered monsters, but it’s very smooth for me on both desktop and phone, so it must not be too intensive. I guess I’ll hear from people if it’s choppy on their device.
I think the nature of our profession as coders is in process of shifting very rapidly, from "write code to do something useful" to "write code to do something useful, better than I could vibe code myself".
Feels like the painful transition when professional photographers started having to differentiate themselves from whatever people could do with their own phone.
On the other hand, as someone who can code in certain domains (web, maps), I could definitely see myself vibe coding as a way to quickly create something in a domain where I have no expertise (eg, Unity).
As for my own work, I just spent a couple hours this afternoon in a back and forth discussion with claude code, asking it to mock up a UI for me before "we" start building it tomorrow. It was just a mock-up, so I didn't require precision, but I was impressed with some tidbits that came along for the ride.
Some things it did without me asking
* Mock data for the lists and pages in json format, so I could easily add records to it for different scenarios
* Working navigation between pages, including modals
* Working progress bars and timers
* Working list sorts and filters
* Toasts for functionality that was beyond the scope of the mock-up ("sending email to author of post" or "banning user")
* Not-half-bad animations and transitions between pages, screens, modals, etc
* A responsive layout that worked better than expected on mobile and desktop
* Some ideas I hadn't considered, that we then expanded upon
I would have mocked this up for a client, but not for myself. It's quite nice to have a working html / javascript / css mockup to play with while I flesh out my own ideas - with a benefit that I actually fully understand the output and can tweak it myself as needed.
A) Lots of useful apps aren't a great fit for a spreadsheet. AI seems to be opening many of those up the same way.
B) Lots of spreadsheets have bugs which cause then to give wildly inaccurate results, which are relied on to make crucial decisions. AI is also repeating this part of the pattern.
If you need it to work correctly all the time, there's still no substitute for expertise - but looking at the state of computing, clearly many people are willing to use things that have obvious, serious bugs.
I can look at someone’s finished pivot table and reproduce it from the data through other means, but any explanation of what a pivot table actually is and does reads like pure gibberish to me.
I wonder if there's gonna be a big reckoning of these "AI-first" companies suddenly people to fix their shitty vibe coded stuff. Ooh, new job title: AI fixer, minimum salary, $1m/yr, with RSU.
- LLMs make writers (for the kind of formulaic crap we ghost-write) unnecessary. We're firing 75% of our writing + editing staff, mostly the writers. Editors will now use LLMs to do the whole thing.
- What do you mean even our best editor and best LLM-user is still taking 90% as long as the total writing + editing process before, and their parallel-work capabilities are down, so throughput is down to a trickle and also every remaining worker is pissed off about how intensely frustrating working with LLMs is so the whole team's grumbling about leaving? Of fuck, we committed (without asking you, naturally, and ignoring a ton of explicit warnings that this was a growing problem) to deadlines!
- Hires back nearly all the writers as contractors
It existed with adobe flash. As much as programmers hated flash, it allowed artists with little technical skills to create awesome mini games.
Hmm, do we have statistics about that? Like, did the profession collapsed?
I wouldn't equal the photo I do (not much) with anything close to a pro. Not even to a good amateur level actually. But that's just me. :D
The "nature of the profession" might already be quite diverse, but that's an interesting remark.
I know this is not the case here and the game is very cool, I was primarily replying to the comment about the new trend.
You also invented a movement control method I have never seen before - please keep making games.
Your very own Steve Jobs Roll Your Own Calculator Construction Set[1].
(It is of course very common to do all sorts of game art using ad hoc parametric stuff like this, I just find the similarity amusing.)
[1] https://www.folklore.org/Calculator_Construction_Set.html
Finally camped by the health and was rewarded with...one health.
Kept hoping the +spread would shoot closer to down.
Aside from that, this might become my new favorite time waster of the week.
"I was pretty happy with the game and ready to share it. But then at the last minute I got another nagging idea in the back of my mind: What if it was somehow more like actual doomscrolling?
It would be easy to get an RSS Feed of headlines from a news site. Could I make them appear in the game as you scroll, in a way that felt integrated with the game?"
I feel like you are forced to get the power upgrades at first to get past the larger roadblocks before the fire wall hits you, but maybe this is avoidable if you're fast enough.
Also, this is the first time I'm genuinely impressed by some LLM coded output, bravo to both you and chatgpt.
I know this is just fun and games, but i cant even start to imagine what the code is like.
About half of my week is spend troubleshooting old legacy code for this or that industrial automation devices and PLCs, the latter being their own animal driven by ladder logic. Regardless if it's some firmware in C, some proprietary app pounded out in some ancient .NET version or just plane old PLC programming, the amount of times I have puzzled over what the heck the original programmer was thinking is mind blowing. Prior to moving into this position, I had no idea what the crufty greybeards meant when they were griping about spaghetti code. Now I do. Stepping back through someone else's work, often done with some kind of set-it-forget-it mindset because, for some reason, people create things and ignore that the environment will eventually change, is the biggest, dumbest headache. So dumb, that about a year ago, I start practicing things like writing firmware from scratch because just starting from zero turned out to be easier than trying to fix someone's old, convoluted mess.
Most humans code crap. Some humans code very, very well and write beautiful, eloquent software as a result. Congrats to those in that crowd, but understand they are the minority. So, when we started feeding a stochastic parrot some code to teach it to do our work for use, we fed it garbage. And what happens when garbage goes in?
My biggest frustration with AI coding tools is that bad engineers are no better at judging the quality of AI code than they are at writing code themselves. So, their output has shot through the roof without an improvement in quality.
The productivity of good engineers has gone up as well but good engineers tend to actually think about what their tools are doing, which slows them down. Bad engineers are now able to output more shit code than ever before.
I feel like I’m watching my company building a house of cards in real time.
Great game! Few impressions of the gameplay - Way easier on laptop compared to mobile; the monsters still spawn at the bottom of the viewport so you can camp at the top much further away on a laptop - The game almost gets easier the more powered up you get. To make it more challenging you could consider speeding up the lava at the top
Yes. MVP: <https://sc.rollc.at/>
https://www.grumpygamer.com/deathbyscrolling5/
Even the fire effect coming after you looks similar.
Another way of decreasing the difficulty is to enlarge the "playfield". I actually did that by chance, because my browser window is usually only half of my 4K monitor, so it's taller than wide. Because the monsters always start off at the bottom, you have much more time to shoot them.
* Yes the game is easier on taller screens where you can see more of the game at once. I considered making gameplay literally identical everywhere by scaling it or capping the size, but this way felt more in the spirit of real doomscrollable content, since web pages are taller on different devices.
* The game is not inspired by Ron Gilbert’s upcoming Death By Scrolling, which I only just learned about and am relieved to see is not the same concept. But I’m a Ron Gilbert fan and can’t wait to play it.
* Yes, you can cheat the Wall of Fire, as people noticed. It’s not constantly present. Once it’s a certain distance behind you (I think 20m), it’s not in memory anymore and won’t chase you. It only shows up again if you remain within the same 10m area (+/- 5m) for 7 seconds or longer. So if you move up and down enough, it won’t come.
* A few people have commented about the scrolling speed and momentum. I actually worked on this, it wasn’t up to the LLM to decide. I had the LLM make a config file with parameters I could tweak for top speed, friction, momentum, acceleration, deceleration, etc., and played with it until I was happy with it on my devices. But I did play on a borrowed device and it had a slightly different feel. So it may feel different on your device or perhaps you just would have preferred different settings. I don’t know if an experienced coder would have written it in a way that is a better experience across all devices to begin with, or if play testing would have helped me identify cross-device differences I could still fix with vibe coding.
* A little more behind-the-scenes because it seems like people here might enjoy more details: I set the sprite baseline for the first monster’s animation at the bottom so it would “bounce” properly, and this made it so the collision box wasn’t placed right because the code kept using the sprite bottom as the center of the collision box. The LLM had a really hard time understanding that the box should match the sprite and I spent a lot of time struggling with it. So for debugging purposes, I made it possible to reveal the collision boxes by typing the ! (exclamation point). I left that in and you’re welcome to toggle that if you feel like it. It also exposes the collision box for webs so I could determine how much they should overlap before getting caught in a web.
* Another debugging thing: I needed to be able to trigger a weapon upgrade manually so there’s a hotkey for that. It’s not too hard to find but shhh no cheating.
* Someone noted that this feels like “the painful transition when professional photographers started having to differentiate themselves from whatever people could do with their own phone.” I was a professional photographer working in NYC during the commoditization of photography and I 100% relate to this. “Why should I pay you $800 for a photo that my nephew can take for $50?” was the sort of thing we heard a lot. Photography hasn’t recovered. I successfully pivoted to video.
* This isn’t my first vibe coded game or my first time on Hacker News. So if you found this interesting, you should sign up for my newsletter: https://ironicsans.ghost.io
Thanks everyone!
Couple constructive criticisms:
- Is there a pause button that I'm missing? If not, add one, I want to actually play this more but I need to be able to take breaks
- The beginning is too easy. I'm biased as I like difficult games that encourage mastery, but I found myself dying early on runs because I was getting impatient with how slow the early progression is
- I haven't played enough yet to know whether monsters eventually come from the top, but eventually they should, it would greatly increase the difficulty curve - Hitboxes feel a little meh, and don't seem to match the sprites
- If there's sound, it wasn't working for me, which leads to...
- Pump up the juice. You've got a great core loop, but you need the juice. One of my favorite game dev talks of all time would be perfect for you to watch and iterate with (https://www.youtube.com/watch?v=AJdEqssNZ-U)
- Monster diversity. Again, maybe I didn't play long enough yet to get to newer monsters, but from what I can tell the only diversity is their looks, health, and approach angle. Monsters that are faster and incentivize different movement, or that shoot back at you, or blow up after shooting so you have to avoid, etc.
I'll stop there, don't want an absolute laundry list. Keep at it, you've got a unique foundation, it'd be a shame not to polish it up!
I played for about 300 monsters and it got boring.
lif•4mo ago