FredrikNoren

6 ms·

1. 1.▲
   by FredrikNoren 2y ago

   Wow that's great! Thank you!
2. 2.▲
   by FredrikNoren 2y ago

   Thanks! That's really interesting stuff, will definitely look into those!
3. 3.▲
   by FredrikNoren 2y ago

   I'm experimenting with an AI assisted world building / story telling tool: https://youtu.be/OGkSI3VfxRU . The idea is to do kind of what Cursor is doing for programming; accelerate the creative process (rather tha
4. 4.▲
   by FredrikNoren 2y ago

   So... I think maybe the same could have been said about writing once upon a time. "What if everyone could write, we'd flood the market with poorly written books!". Well we went from no one being able to write to almost everyo
5. 5.▲
   by FredrikNoren 2y ago

   Nope, that's unrelated.
6. 6.▲
   by FredrikNoren 2y ago

   So it's hard to give a number on how robust it is, but yes, simply telling GPT that "this doesn't work" or "try again" does work a lot of the time. We're trying to give GPT as much context as possible to h
7. 7.▲
   by FredrikNoren 2y ago

   Thanks! We're using a service called Meshy ( https://www.meshy.ai ) to generate 3d models.
8. 8.▲
   by FredrikNoren 2y ago

   Thank you!
9. 9.▲
   by FredrikNoren 2y ago

   I'd say it's something in that direction (GameMaker) yeah. One big difference (other than the AI part) is that both building and playing is multiplayer by default. (Though of course, you can choose to make things single player and
10. 10.▲
    by FredrikNoren 2y ago

    We haven't looked into exactly how this will be handled yet, but I believe this would fall under the same laws and restrictions as any other online platform (YouTube, TikTok etc.), it's just that the production tools are different
11. 11.▲
    by FredrikNoren 2y ago

    Thanks! Oh yeah that would be really interesting to explore! We haven't done much with LLMs on the play side yet, but we know that there's so much that can be done there, that can open up for totally new types of games.
12. 12.▲
    by FredrikNoren 2y ago

    So far we've focused 100% on the creation side of things, but we're absolutely interested in helping people on the marketing side too. Something we've discussed here is; since it's an online platform, could it be possibl
13. 13.▲
    Building an AI game studio: what we've learned so far(braindump.me)

    287 points by FredrikNoren 2y ago|297 comments
14. 14.▲
    by FredrikNoren 2y ago

    Hi HN. We want to make it possible for anyone to build games without first learning to code or make art, by giving them access to an “AI game studio”. It’s early days, but here’s what we’ve learned so far building this product. Would love t
15. 15.▲
    by FredrikNoren 3y ago

    Well, only a few people have tried it out so around $30 so far (haven't checked exactly how many but less than 30). The prompting is very unoptimized as well, so I'm sure it's possible to get that down.
16. 16.▲
    by FredrikNoren 3y ago

    Hm yeah I can see your project; it looks like the players were created outside of the map. If you click-drag the map a bit you'll see them in the lower right corner. I'll see if I can give it a better understanding of the size of
17. 17.▲
    Show HN: Prompty, build games with GPT (early prototype)(prompty-58a21.web.app)

    11 points by FredrikNoren 3y ago|4 comments
18. 18.▲
    by FredrikNoren 3y ago

    Thanks for detailed feedback!
19. 19.▲
    by FredrikNoren 3y ago

    So we view the browser as one distribution channel among many. We want Ambient to run everywhere. The first platforms we built out Ambient for were actually desktop native, we've since then also added the web support. That said, we do
20. 20.▲
    by FredrikNoren 3y ago

    Oh yeah, absolutely agree, we're not stopping at 60 ms. Client side prediction is planned (see Philpax answer https://news.ycombinator.com/item?id=37767359 and our netcode roadmap: https://github.com/Am
21. 21.▲
    by FredrikNoren 3y ago

    Hm yeah, we need to fix the detection as part of instantiation, which happens in the engine. I'm adding a "Try anyway" button for you in the meantime, it should be up in a couple of minutes.
22. 22.▲
    by FredrikNoren 3y ago

    Yeah, it's part of the problem, but that would normally only give you ~60 - 100 ms lag (at least if you're on a decent connection, close-ish to the server). But we're experiencing another bug where WebTransport doesn't g
23. 23.▲
    by FredrikNoren 3y ago

    So if you browse to any game on you should see a download button to the left. Sorry to hear that the website doesn't feel good! Which parts of it specifically are you referring to?
24. 24.▲
    by FredrikNoren 3y ago

    It's built on top of WebTransport, which is also quite new. But yeah, we're also seeing the input lag issue here and we're on it, will be fixed soon! When it comes to loading resources ahead of time: this is something devs ca
25. 25.▲
    by FredrikNoren 3y ago

    Pros: It's basically just a more modern API, so we can do a lot more (compute shaders for instance). There's a lot we can do on the performance side with it to make it faster than what we could do with WebGL. Cons: It's not y
26. 26.▲
    by FredrikNoren 3y ago

    It's fixed on the website now as well (it'll show you a nice warning screen on non-webgpu browsers, instead of just loading forever)
27. 27.▲
    by FredrikNoren 3y ago

    Which version of chromium are you on? WebGPU is really new so it may require a browser update.
28. 28.▲
    by FredrikNoren 3y ago

    Haha indeed :)
29. 29.▲
    by FredrikNoren 4y ago

    We're actually only three full time devs :) But we're hiring! See the careers section on the website.
30. 30.▲
    by FredrikNoren 4y ago

    Hi, Ambient CPTO here; We're using wgpu (a rust crate), which supports a number of different backends, including Vulcan and WebGL (and Directx, WebGPU, Metal). We're VC backed, which is how we finance the project right now. Long t