3 ms·

Turns out making a game, even with full help of the LLM power is still a massive effort most people won’t go through. Even simpler but non-trivial programs req

by yoz-y 4mo ago

Turns out making a game, even with full help of the LLM power is still a massive effort most people won’t go through. Even simpler but non-trivial programs require a lot of back and forth. So in the end it will be the same kind of motivated people that will be able to produce something good. We’re nowhere near “Claude, build me GTA6”

AndrewOMartin 4mo ago
We might get a powerful enough model to run "Claude, build me GTA6" before GTA6.
- tripledry 4mo ago
  Highly doubt it.
  - pennaMan 4mo ago
    someday we will have models that can resolve physics to such degree to predict the future with surgical accuracy and when someone says "maybe models will become advanced enough to create a whole other universe from scratch" you will be there saying "highly doubt it"
    - grebc 4mo ago
      Someday.
    - owebmaster 4mo ago
      This reminds me of a boss I had 20 years ago that said devs were going to be replaced soon. Maybe he was 20 years early or maybe it's not happening now too.
      - pennaMan 4mo ago
        Please note that I did not imply devs won't exist in my contrived fantasy. Who do you think will be operating those mythological tools?:)
  - adornKey 4mo ago
    Hm. Nowadays a lot of games look alike. They use some 3D-Engine and most of the work is in the 3D-modelling and writing some interesting scripts. For modelling and scripting I think we're not far away. A lot of games just reused old historical stories or fiction and a lot of stories feel like cheap soap operas. As soon as an AI can separate the good from the bad scrips it'll be mostly done.
    - kakacik 4mo ago
      Clearly you don't work in game development, this comment put a smile on my face. Although I expected a bit more from hacker news crowd
      - whizzter 4mo ago
        You overvalue the HN crowd (or undervalue the AI hype-machine) considering you're downvoted and GP is upvoted (Another gamedev here smiling at GP's comment).
        gafferongames 4mo ago
        Yet another classic unskilled but unaware of it from the hacker news crowd (another professional gamedev here).
    - Cthulhu_ 4mo ago
      I'm sure AI will be used in asset generation but it'll be in deployed similarly as procedural generation of mass assets like trees and npcs. I don't yet see it used for characters as they quickly become kind of generic / predictable.
  - inigyou 4mo ago
    It's not hard if you just imagine a slop re-skin of any other GTA of similar game.
  - kobalsky 4mo ago
    Did you miss the sarcasm or has GTA6 become some sort of anomalous memetic agent that makes anyone who tries to work on it not be able to finish it regardless if it's human or AI?
    - tripledry 4mo ago
      Both probably, I don't usually understand sarcasm in written text unless it's explicitly stated.
uludag 4mo ago
Here's an example of this regarding the recent indie hit "Mina the Hollower." They were recently interviewed and asked the question "What is Yacht Club's stance on AI and has AI been used in the production of Mina the Hollower at all?" to which their reply was "We all got caught up in AI fever like the rest of the world, but we didn't find it was very effective for what we're doing. Maybe our work just isn't that generic! We've found some ways it can help... like Google or a thesaurus, but it hasn't affected what's in our game." [1] It's hard to immagine that even the best AI model would result in anything better than a marginal reduction in release timeline. Like maybe for such projects one could spend $X00,000 worth of tokens for maybe like a %single-digit-percent reduction in time to release. Marginally good, maybe even project saving, but not any larger a paradigm shift than Unity was. [1] https://www.gamereactor.eu/mina-the-hollower-interview-discussing-the-delay-length-ai-inspirations-and-secrets-with-yacht-club-1718863/ https://www.gamereactor.eu/mina-the-hollower-interview-discu...
- Zarathruster 4mo ago
  The conversation around AI in games is such a polluted mess of confusion and bad faith. I would hate to have to answer such an interview question. Recently I came across a game on Steam that was getting review-bombed because the devs admitted to using AI on an entirely different game. By "using AI" I mean they admitted to using Cursor of all things. Not assets, just a coding assistant. If we're going to get backlash for stuff as stupid as this, it's probably best to just keep one's entirely mouth shut about all things AI.
- clates 4mo ago
  > "What is Yacht Club's stance on AI and has AI been used in the production of Mina the Hollower at all?" to which their reply was "We all got caught up in AI fever like the rest of the world, but we didn't find it was very effective for what we're doing. Maybe our work just isn't that generic! We've found some ways it can help... like Google or a thesaurus, but it hasn't affected what's in our game." [1] Can you see how this intentionally coy and evasive answer was made specifically to not say "No, we didn't use AI." but still sound like "No, we didn't use AI." so that people wouldn't immeidately bust out pitchforks and start review bombing their game before it takes off?

by yoz-y 4mo ago

AndrewOMartin 4mo ago

We might get a powerful enough model to run "Claude, build me GTA6" before GTA6.

tripledry 4mo ago
Highly doubt it.
- pennaMan 4mo ago
  someday we will have models that can resolve physics to such degree to predict the future with surgical accuracy and when someone says "maybe models will become advanced enough to create a whole other universe from scratch" you will be there saying "highly doubt it"
  - grebc 4mo ago
    Someday.
  - owebmaster 4mo ago
    This reminds me of a boss I had 20 years ago that said devs were going to be replaced soon. Maybe he was 20 years early or maybe it's not happening now too.
    - pennaMan 4mo ago
      Please note that I did not imply devs won't exist in my contrived fantasy. Who do you think will be operating those mythological tools?:)
- adornKey 4mo ago
  Hm. Nowadays a lot of games look alike. They use some 3D-Engine and most of the work is in the 3D-modelling and writing some interesting scripts. For modelling and scripting I think we're not far away. A lot of games just reused old historical stories or fiction and a lot of stories feel like cheap soap operas. As soon as an AI can separate the good from the bad scrips it'll be mostly done.
  - kakacik 4mo ago
    Clearly you don't work in game development, this comment put a smile on my face. Although I expected a bit more from hacker news crowd
    - whizzter 4mo ago
      You overvalue the HN crowd (or undervalue the AI hype-machine) considering you're downvoted and GP is upvoted (Another gamedev here smiling at GP's comment).
      - gafferongames 4mo ago
        Yet another classic unskilled but unaware of it from the hacker news crowd (another professional gamedev here).
  - Cthulhu_ 4mo ago
    I'm sure AI will be used in asset generation but it'll be in deployed similarly as procedural generation of mass assets like trees and npcs. I don't yet see it used for characters as they quickly become kind of generic / predictable.
- inigyou 4mo ago
  It's not hard if you just imagine a slop re-skin of any other GTA of similar game.
- kobalsky 4mo ago
  Did you miss the sarcasm or has GTA6 become some sort of anomalous memetic agent that makes anyone who tries to work on it not be able to finish it regardless if it's human or AI?
  - tripledry 4mo ago
    Both probably, I don't usually understand sarcasm in written text unless it's explicitly stated.

uludag 4mo ago

Here's an example of this regarding the recent indie hit "Mina the Hollower." They were recently interviewed and asked the question "What is Yacht Club's stance on AI and has AI been used in the production of Mina the Hollower at all?" to which their reply was "We all got caught up in AI fever like the rest of the world, but we didn't find it was very effective for what we're doing. Maybe our work just isn't that generic! We've found some ways it can help... like Google or a thesaurus, but it hasn't affected what's in our game." [1] It's hard to immagine that even the best AI model would result in anything better than a marginal reduction in release timeline. Like maybe for such projects one could spend $X00,000 worth of tokens for maybe like a %single-digit-percent reduction in time to release. Marginally good, maybe even project saving, but not any larger a paradigm shift than Unity was. [1] https://www.gamereactor.eu/mina-the-hollower-interview-discussing-the-delay-length-ai-inspirations-and-secrets-with-yacht-club-1718863/ https://www.gamereactor.eu/mina-the-hollower-interview-discu...

Zarathruster 4mo ago
The conversation around AI in games is such a polluted mess of confusion and bad faith. I would hate to have to answer such an interview question. Recently I came across a game on Steam that was getting review-bombed because the devs admitted to using AI on an entirely different game. By "using AI" I mean they admitted to using Cursor of all things. Not assets, just a coding assistant. If we're going to get backlash for stuff as stupid as this, it's probably best to just keep one's entirely mouth shut about all things AI.
clates 4mo ago
> "What is Yacht Club's stance on AI and has AI been used in the production of Mina the Hollower at all?" to which their reply was "We all got caught up in AI fever like the rest of the world, but we didn't find it was very effective for what we're doing. Maybe our work just isn't that generic! We've found some ways it can help... like Google or a thesaurus, but it hasn't affected what's in our game." [1] Can you see how this intentionally coy and evasive answer was made specifically to not say "No, we didn't use AI." but still sound like "No, we didn't use AI." so that people wouldn't immeidately bust out pitchforks and start review bombing their game before it takes off?