Article
files/splash/LINEUP

Game AI Roundup Week #19 2008: 7 Stories, 1 Quote, 2 Videos

Novack on May 11, 2008

Weekends at AiGameDev.com are dedicated to rounding up smart links from the web relating to artificial intelligence and game development. This was a more quiet week but you’ll still find a few great blog posts, articles and videos. Remember, there’s also lots of great content to be found in the forums here! (All you have to do is introduce yourself.) Also don’t forget the Twitter account for random thoughts!

This post is brought to you by Novack and Alex Champandard. If you have any news or tips for next week, be sure to email them in to editors at AiGameDev.com. Remember there’s a mini-blog over at news.AiGameDev.com (RSS) with game AI news from the web as it happens.

Why Game AI Still Sucks

Sam Larbi made a post on his blog, in response to Dave Mark’s article “Sharing the Sandbox: Can We Improve on GTA’s Playmates?” here on AiGameDev.com.

“Dave Mark raises some interesting questions about artificial intelligence in games over at AiGameDev.com. First, he explains that although we’re seeing more and better AI in games, a common complaint heard from gamers runs along the lines of “why can’t they combine such and such AI feature from game X in game Y.” Then, Dave poses the questions for developers to answer”

Starcraft 2’s AI Does Not Cheat Like Before

In a recent interview with 1Up editor-in-chief James Mielke, Chris Sigaty (StarCraft 2 lead producer) talked about the recent developments that Blizzard has made with the game, specifically the game’s AI and the Zerg Swarm in the long awaited RTS. In particular, you’ll hear one bit about how the AI is now simulated in the same way as human players — including real fog of war.



Fighting AI Training

Billy Hewlett, a DMA UCLA student, comments on his Final Project Game Idea: an AI Training/Fighting Game. The series of articles on his progress is a very interesting reading as well.

“The idea with this multiplayer turn based fighting game is that you teach an AI how to fight your opponent. The game is a 2D top down Zelda like combat game. In round 1, you fight an easy AI opponent in the fighting game. As you fight your AI opponent, you are also training it to fight your human opponent. In round 2, your opponent fights a more difficult AI opponent, that has learned from your round 1 training. Similarly, in round 2 you fight a slightly more difficult AI opponent who has received training from your human opponent. Ideally, the game is networked, so that each round can occur simultaneously. Your AIs get saved to disk every turn, so you can fight against different human players and start both AIs at round 5 if each player has progressed to that point.”

AIIDE 2008 Dates Announced



The organizers of the AIIDE conference have revealed the dates for this year’s event about game AI research: it will take place in October at Stanford:

“Taking place at Stanford University in the heart of Silicon Valley, California between October 22nd and 24th, AIIDE covers the application of artificial intelligence to games and entertainment, featuring talks about new developments in the AI field and experiences in deploying AI in commercial products.”

GTA IV AI Expectations



AI Panic! published an article entitled “AI in GTA IV: Nothing Spectacular” where the actual AI in the game is compared with what the player’s expect from the AI.

“I’ve written about AI in video games and their real-life impact a while ago, with the conclusion that it is quite basic and does not really contain “intelligence”. Now with all the hype around the newest version of the Grand Theft Auto series, GTA IV, and a development budget of about $100 million, lets take a closer look if anything has changed.”

Funny Quote of the Week

The winner of this week is Stephen Totilo from MTV Multiplayer, with his review of GTA 4.

“Nothing feels more like New York gameplay to me in GTA IV than stealing a car in front of a gathering of people who can’t even be bothered to turn their heads. I don’t know if the artificial intelligence was programmed with that intent, but, let me tell you, that is a New York vibe.”

The Making of Leaf - Part 2

The game developer and lecturer (and a regular in the AiGameDev.com forums), Phil Carlisle, continues on his blog series “The making of LEAF” talking about his efforts to build a village with over 10k inhabitants.

“This week, I’m going to cover the starting point for fulfilling the needs we identified in part 1. Specifically, we’re going to look at the Behavior Tree structure that will form the starting point for developing the AI interactions.”

Procedural Storytelling



On his blog Game of Design, Dan Kline, Game AI Programmer and Designer wrote about Left 4 Dead’s AI Director which handles the high-level story and gameplay flow of the game.

“Dynamic spawning is the first step towards the procedural storytelling I’ve been working on for the last couple of years. Finally some of this stuff is getting published. Yes, dynamic procedural storytelling might ultimately never work. But, if you’ve played a pen and paper RPG you know it probably can.”

Videos of AGI ‘08



In other news, organizers of the AGI-08 Conference (which took place in march this year) uploaded the videos and papers of many conference presentations — including a talk by John Laid on SOAR and a few less technical talks on virtual worlds.

“Papers linked. Adding videos and transcripts. Last updated Apr. 30, 2008

  • Introduction

  • Overview of AGI Research

  • Architecture of AGI Systems

  • Language and Cognition

  • Reasoning

  • Learning

  • Virtually Embodied AI

  • Catalyzing the Coming AGI Renaissance

  • Neural Network and Brain Modeling

Moding AI

Is still amazing what the users can do, given some simple tools to play with!



“Using a combination of ai_goal_actbusy and scripted_sequence, I sucessfully got an NPC to use an elavator intended for players. The simple elevator had three buttons, on to call it to the second floor, one to call it to the first floor, and one to make it go up or down depending on where it was. The NPC would first check the elevator to see if it was at their floor or not. If it was, they would enter it and press the button to move it to the bottom floor. If it wasn’t, they would press the button to call it, wait, then they would enter it.”



Stay tuned next week for more smart links from around the web!

Discussion 1 Comments

alexjc on May 13th, 2008

Thanks for holding the fort this weekend, Marcos! It's been a busy few days :-) Anyway, my edits are in now... Alex

If you'd like to add a comment or question on this page, simply log-in to the site. You can create an account from the sign-up page if necessary... It takes less than a minute!