Hacker News

Top stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms (github.com/firelex)
    33comments
  2. Pirating the Pirates (mubi.com)
    181comments
  3. 12,000-year-old Göbeklitepe burials explain scattered bones (archaeologymag.com)
    7comments
  4. MicroLLM Lab – Try 7 tiny LLM's in the browser (stateofutopia.com)
    41comments
  5. World Labs Is Joining AMD (worldlabs.ai)
    44comments
  6. Scientists solve 1840s space weather mystery (arstechnica.com)
    20comments
  7. Hijacking the PS5's RTMP stream (yashgarg.dev)
    54comments
  8. What reversing, modernising old games tells us about the economic impact of AI (isfine.org)
    9comments
  9. Who should be held accountable when an AI Agent (accidentally) acts maliciously? (greenpants.net)
    5comments
  10. Joseph Szabo’s pictures of American adolescents (newyorker.com)
    30comments
  11. It's Time to Investigate the AI Labs (calnewport.com)
    45comments
  12. Kids turned low-traffic NPR Spotify comments into a secret group chat (thisamericanlife.org)
    139comments
  13. Parley: Federated, decentralised chat that speaks plain IRC (mills.io)
    145comments
  14. Sonnet 5.5 (anthropic.com)
    336comments
  15. Best of British Design (best-of-british-design.vercel.app)
    21comments
  16. Flock Wants the Most Detailed Map of Its Surveillance Cameras Taken Offline (theintercept.com)
    28comments
  17. Pacing the Frontier is not the actual goal for AI labs (lesswrong.com)
    48comments
  18. Cf: The Agentic CLI for the Cloudflare API (cloudflare.com)
    41comments
  19. 3D necroprinting: Leveraging biotic material as the nozzle for 3D printing (science.org)
    —discuss
  20. First Steps of the PLC Organization – Independent Public Ledger of Credentials (plcred.org)
    18comments
  21. Nvidia wants to put a watchdog chip next to every AI agent (cnbc.com)
    122comments
  22. Launch HN: Vespper (YC F24) – SOTA Docx MCP (vespper.com)
    8comments
  23. I made a visual workspace for AI Automations (biom.dev)
    22comments
  24. Show HN: Destroy Any Website with Stickman (spritefusion.com)
    25comments
  25. Who wrote Elizabeth I's most scathing letters? (smithsonianmag.com)
    18comments
  26. What heraldry and Japanese mon can teach about visual-identity generators (benovermyer.com)
    19comments
  27. Updated Google Maps shows destruction of the city of Rafah (twitter.com/aliabunimah)
    35comments
  28. Does Reddit have an astroturfing problem? What the data suggests (petervijeh.com)
    79comments
  29. MongoDB CEO resigns to join Meta (reuters.com)
    252comments
  30. GrapheneOS – When an app is slow (wirelessmoves.com)
    38comments

What reversing, modernising old games tells us about the economic impact of AI

23 pointsby 1d agothis.os.isfine.org
9 comments
1d agoHN ↗

Cross posting from another thread: 17 games fully or partially reversed and ported to the web by one very capable dude in a few weeks with frontier and open weight models. He also did this thing where he “spliced” different game concepts and turned WoW into an FPS: https://news.ycombinator.com/item?id=49652089

In case you mistake him for an “AI booster”, you should see this guy’s other posts, he is extremely critical of the AI frontier labs and Tech Bros and this whole AI bubble.

But he is also a very experienced game developer, and under his control the latest models have pulled off these crazy feats. What TFA is the analysis of a domain expert and what it means for the industry.

37m agoHN ↗

Just so you know, the other submission you linked here got killed (i.e. is not visible without showdead on).

35m agoHN ↗

Why was it killed? That sounds very interesting.

Does anybody have a link to it? I can't find it on the blog.

13m agoHN ↗

Whoa TIL. Too much low-quality stuff being cross-posted from LinkedIn?

11m agoHN ↗

Lots of would-be business influencer types post (or posted? I've never paid much attention to it) on there. There are a number of other sites that are also auto-killed by the system. Sometimes those submissions will come back to life, but it has to be found by a mod or vouched for by enough users.

6m agoHN ↗

I would like to know: has anyone actually played any of these games he's "modernised"? I would expect them to be filled to the brim with papercuts and, perhaps, rusty nails. LLMs are no doubt capable of translating code when it has a perfect, byte matching spec to validate aganst, which is essentially the most rigorous test harness it is possible to have. They don't even need to be good at it; they just need to be persistent, correcting one error after another for 24 hours a day to reach the destination.

However, I have not been impressed with their capability for gamedev beyond that. I do have a $200 codex sub, and I strive to use the full budget every week / every reset. Much of this usage is vomited into a trash bin exploring what works and what does not, and there is a lot more that does not work than does. The biggest benefit is that you can take your attention off something and have it work in the background, obviously, but the longer it goes working on something that does not have a 100% accurate test spec to adhere to, the more it goes off the rails no matter how much time I waste "harness engineering", with it notably contradicting concrete instructions as context size increases. Even on single, extremely targeted tasks, they are of limited value, with max reasoning often taking 10~30 minutes to complete a task I could do in minutes by hand, with significantly worse and less maintainable code, and often introducing edge case bugs or unusable UI.

I would not be surprised if this is just yet another showcase that looks impressive at surface level, as LLMs have been benchmaxxed to do, but falls apart as soon as you look behind the facade. That said, feel free to correct me if I'm wrong and you think these modernisations are masterpieces. I don't really have any desire to play them myself, but maybe I'm underestmating them.