Game AI Roundup Week #22 2008: 9 Stories, 5 Jobs, 2 Events, 1 Video

Novack on June 3, 2008

When there are no technical glitches, weekends at are dedicated to rounding up links from the web relating to game AI. This post is a little late, but there’s been lots of activity this week; several jobs, a workshop in Paris and solid articles and blog posts for you to read. Remember, there’s also lots of great content to be found in the forums here! (All you have to do is introduce yourself.)

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

Don’t Follow the Shortest Path!

Christer Ericson from wrote an article on his perspective on A* — in his usual style of cutting straight to the matter:

“I wish AI programmers all over the world would think more about the heuristic function, if they haven’t already, and stop being totally obsessed with shortest paths. Instead, use nonadmissible heuristics and cut your search effort in half!”

Be sure to read the comments for our take on the subject, and more explanations from Christer.

Far Cry 2 AI Impressions

Nick Breckon from Shacknews wrote a preview on Far Cry 2 where he remarks about new open-world-type AI features:

“Enemies are purely dynamic, never spawned. Instead, they simply exist in the world, sometimes called in from other areas by signal flares. Amancio recalled one instance in testing where, after reinforcements were sent to finish him off, he had five enemy trucks on his tail, a record for testing. Needless to say, he died soon after.”

AI Fighting Game Now with “Reinforcement Learning”

Billy Hewlett continues his blog series describing the process around his univertity Final Project.

In response to gameplay testing feedback [that it was difficult to train the AI and that the non-active player got bored], I’ve added a teaching mechanism for the inactive player. The inactive player controls a small “teacher” version of their character in the lower corner of the screen. The teacher can either cheer their AI on or admonish their AI and tell them what they should be doing. The cheering raises the AI’s morale meter [which currently does nothing, but soon will have some flashy effect when full]. Undiscriminating cheering raises morale slightly, but cheering after the AI accomplishes something (like hitting the other player or dodging an arrow) raises morale by a significant amount.

Artificial Idiocy

Next Generation featured an article where the author makes a very interesting (and lately recurrent) question:

Witless enemies are an oft-cited failing of videogames, but while AI designers are forever finding ways to make more formidable opponents, is smarter always better?

The 12th International Computer Games Conference

Here’s a call for papers from the U.S.:

“You are invited to submit a paper on any topic related to computer games design, development and education, and particularly papers covering the use of AI for modelling and programming “believable characters”, mobile games, multiplayer on-line, educational and serious games. We strongly encourage the submission of papers related to the design and the experiments of original type of games, in particular in the fields of serious games, ubiquitous, mobile, cross media platforms and massively multiplayer on line games.”

Delatorian Trails

We already saw enemy NPCs pursuing the player, but following him… by trails? Promising additions for the next Alone in the Dark AI are commented on Play, on a preview by Luke Guttridge.

“Artificial intelligence is also receiving plenty of attention. You’ll face both human and “possessed” enemies apparently, and there’s a few interesting touches like foes being able to see you through glass and follow trails of blood that should make the game more interesting.”

Then again, this is starting to sound like the simplest kind of bread-crumb AI that was available in Quake 1/2.

Jobs of the week

Plenty of open positions this days, take a look!

Ubisoft is looking for a Senior AI Engineer for the Shanghai Studio

Job Description:

  • Research and implementation of core character simulation and artificial intelligence for AAA games on Next Generation Game Consoles

  • Develop editor tools for level designer to implement game design related to character AI

Raven Software looks for a Senior AI Programmer.

Job Description:

Raven Software, developer of numerous award winning games, including Quake IV and Marvel: Ultimate Alliance is seeking a senior AI programmer.

THQ, Kaos Studios needs a Senior AI Programmer:

Job Responsibilities:

  • Design and Develop AI and Game Play systems for the game engine within schedule

  • Participates in Design and Code reviews

  • Create systems that achieve the game design goals

  • Create technology that can be re-used and extended in the future

  • Work with other team members to identify, define and solve problems

  • Work with external technology as needed

  • Other duties as assigned

Surreal Software Inc. goes for a Senior AI Programmer.

Surreal Software Inc. (a subsidiary of Midway and the studio which created The Suffering and Drakan) has an opening for a Senior Artificial Intelligence programmer to help develop the AI source base for This is Vegas (our new open-world game for Xbox 360, PS3 and PC) and future titles.

Lucasfilm Animation Singapore is looking for a Sub Lead AI Engineer.

The Sub lead AI Programmer will be responsible for designing an automated system to control the behaviors, short and long term goals, and reasoning of AI in games.

Beyond the Sands of Time

A new Prince of Persia is on its way, and the most notorious addition is an AI female partner in the vein of Alyx, but apparently with much more interaction abilities.

“The Sands of Time gameplay mechanic was hard to give up,” says Mattes, “but the replacement will surpass it.” And what is this amazing new gameplay element? Her name is “Elika”: the new AI-driven support character who will be stepping up to fill Farah’s shoes as the game’s primary love interest/sidekick/action partner.”

There’s lots of potential here, but as always with such difficult problems, it could go horribly wrong.

Nvidia Go Natural

After the free release of Havok for commercial use on PC, the GPU manufacturer was expected to react. Now is announcing a partnership of Ageia products with NaturalMotion. Gamasutra talked with with NaturalMotion CEO Torsten Reil to discuss the partnership and what we could expect from the integrated tools.

Gamasutra: It also seems quite difficult that by the very nature of a physics-based ragdolling, you don’t know what position they’re going to be in. If you start up an animation again, they could be upside-down or something.

NaturalMotion: That’s true, but again, we have quite sophisticated algorithms, because we’re doing so much of that stuff with Euphoria, basically. It’s running in real-time, but Euphoria goes beyond all that, again, because it has behaviors running on top of it.”

BrainWorks: Little Lies

Ted Vessenes wrote yet another very good post on his blog BrainWorks.

As I wrote earlier, I see strong parallels between designing AI and being a parent. There’s a lot to be said for the “lies” that we as AI designers tell our AI: “No really, you can have perfect information about the entire world. And if you have to spend a long time thinking, that’s okay. The entire world will stop and wait for you to decide.” Whenever we let our AI cheat at the game, look up information it shouldn’t know, or otherwise do things a player can’t do, we are in some sense lying to our AI about what reality is like. And the price we pay is this:

We live in fear that our AI will encounter a game situation where the player will know the AI has cheated.

Seeing Eye AI

Triple Entente Radio posted a review on Burnout Paradise (X360); the article denotes another use for a game AI: emergent guidance for the player.

“The game’s AI tends to stick to the main roads, and with no huge arrow or closed roads to show you where to go, you’re often left lost and confused if you don’t follow the computer-controlled cars to the finish line and boost ahead last minute. And while you might have found a great hidden road that goes above traffic, watch out that it doesn’t lead you away from the finish line instead of towards it.”

Paris Game AI Workshop 2008 on June 25th

You’ll be hearing more about this over the course of the week, but we’re organizing our very own event in France in a few weeks:

“The largest online community for game AI has teamed up with one of the foremost research institutes in Europe to organize their first workshop in Paris entirely dedicated to game AI. and the Université Pierre et Marie Curie (Paris 6) LIP6 are proud to announce the Paris Game AI Workshop ‘08, set to take place on Wednesday 25th of June. The workshop is scheduled the day after the Paris GDC, with more specialized content for developers interested in focusing on artificial intelligence for games.”

Space is limited and places are going very fast, so be sure to sign up as soon as possible.

10,000 character crowd

This week’s video is a previous prototype of Kynogon’s technology, showing a very large crowd, managed by its Kynapse AI.

“This is a demo of the Kynapse AI engine from Kynogon. 10,000 characters wander in a large terrain including 2 villages. Kynapse is running on a separate thread.”

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

Discussion 1 Comments

Andrew on June 4th, 2008

While Kynogon’s stuff is cool, I want to see a full rendered and optimised demo - "characters" are never blue blocks in a world with some grey blocks in. I wonder how it works well with LOD and suchlike, where the map is split up into separate areas. :)

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!