I have kinda entered a vortex of browsing through other people's GDC coverage - especially on the sessions that I could not attend. Note that I don't necessarily agree with everything that people have posted here - I'm just including them so people can have a broader picture. Here's a partial list of (loosely) AI-related stuff that I have found so far:

GDC: Storytelling in Bioshock (Not really AI, but interesting)
GDC: Rules of Engagement
GDC: Rules of Engagement Part 2
GDC: A Q&A With Sid Meier (Not really AI... but it's Sid!)
GDC: Creating a Character in Uncharted (animation AI)
GDC: Creating believable crowds in Assassin’s Creed (group behavior and many units)
GDC08 Notes - Streaming Open World Pathfinding (Obviously pathfinding)

A thread at Game/AI where Jeff Orkin (F.E.A.R. AI mastermind) asked what we all saw at GDC it made for an interesting AI discussion.

Labels: animation, Assasin's Creed, BioShock, crowd simulation, GDC, GDC 2008, group behavior, Jeff Orkin, pathfinding, Sid Meier

On Thursday, I attended Ray Kurzweil's keynote entitled "The Next 20 Years of Gaming". For reference, here's the GDC description of the session:

The paradigm shift rate is now doubling every decade, so the 21st century will see 20,000 years of progress at today’s rate. Computation, communication, biological technologies (for example, DNA sequencing), brain scanning, knowledge of the human brain, and human knowledge in general are all accelerating at an even faster pace, generally doubting price-performance, capacity, and bandwidth every year. By 2020, full-immersion virtual reality will be a vast playground of compelling environments and experiences. Initially VR will have benefits in terms of enabling communications with others in engaging ways over long distances and featuring a great variety of environments from which to choose. By the late 2020s virtual environments will be indistinguishable from real reality and will involve all of the senses, as well as neurological correlation of emotions. As we enter the 2030s there won’t be a clear distinction between human and machine, between real and virtual reality, or between work and play. Intelligent nanorobots will be deeply integrated in the environment, our bodies and our brains, providing full-immersion virtual reality incorporating all of the senses, experiences "beaming," and enhanced human intelligence.

The session, while inspiring and informational, seemed to be one that he has given numerous times before - and managed to insert a few token comments here or there that loosely linked it to the game industry. That was rather distracting at times.

His main theme, as usual, was the expansion of no only the processing power (a la Moore's Law) but the "processing power" of the human mind and society as a whole. From a mathematical standpoint, I loved his use of logarithmic graphs and even double logarithmic graphs to show what amounts to constant rates of change as straight lines. He applied this to so many different things, e.g. the growth of life and intelligence, the power of supercomputing, the power over cost of computing, etc.

The crux of this issue is that people have a tendency to think linearly rather than exponentially and especially logarithmically. That means that we tend to mis-project future trends. One way that this hurts us in the technology arena is that, given production times that are getting to 3 or 4 years (Duke Nukem Forever is an outlier), we tend to undershoot the capabilities of the systems that are available by the time that we release our products.

The endpoint, of course, is his much publicized "countdown to singularity" wherein our computing power will match our mental capabilities. A couple of key predictions in this arena (which are kinda creepy in a way):

2010: Computers disappear

• Images written directly to our retinas
• Ubiquitous high bandwidth connection to the Internet at all times
• Electronics so tiny it's embedded in the environment, our clothing, our eyeglasses
• Full immersion visual-auditory virtual reality
• Augmented real reality
• Interaction with virtual personalities as a primary interface

(A great quote: "Real reality will continue to be irksome for a few years.")

• 2029: An intimate merger
  $1,000 of computation = 1,000 times the human brain
• Reverse engineering of the human brain completed
• Computers pass the Turing test
• Nonbiological intelligence combines
• the subtlety and pattern recognition strength of human intelligence, with
  the speed, memory, and knowledge sharing of machine intelligence
• Nonbiological will continue to grow exponentially whereas biological intelligence is effectively fixed

One entire section of his talk was dedicated to the advances and future of nanobiology. While this seemed to be a lull in the relevancy to the game industry, there was a connection there. He actually said that, due to nanotechnology and nanobiology, we would eventually be able to reprogram our bodies the way we reprogram our games. After spending Monday and Tuesday at the Serious Games Summit where the theme was taking game and game programming technology beyond the entertainment world, I couldn't help but think that some of the techniques, especially those related to AI, would map over into some of the genetic and biological applications that he was talking about. Interestingly, this is somewhat related to things I have been reading in the book "A Beautiful Math: John Nash, Game Theory, and the Modern Quest for a Code of Nature".

One of the main interests for me and the people that I attended with was that he was often speaking largely to the AI programmers. One of his comments was that "AI is the next frontier in passing the "uncanny valley" such as it is in games. While hardly a news flash to the game world, I hope that made our stock as AI designers and programmers go up somewhat.

All in all, it was an inspiring speech if not directly relevant to today's game world... but there I go thinking linearly again. I can tell you, however, that his keynote kept coming up in roundtables, conversations, and at the AI Programmers dinner that week - every time a technique of handling prohibitively large numbers of calculations came up, his name was invoked as giving us hope that we could soon be able to handle it.

For more on Kurzweil and all of the above, check out the web site KurzeilAI.net. You can also get his slides from the keynote (9 MB .ppt file) which is where the above shots and images come from.

Labels: GDC, GDC 2008, nanotechnology, Ray Kurzweil, uncanny valley, virtual reality

At the GDC in San Francisco this past week, I had the opportunity to spend quite a bit of time with John Abercrombie of 2K Boston and the AI lead on Bioshock. It had been a heck of a week for the entire Bioshock team since their title had not only won many awards and accolades prior to GDC but also had a great showing at the Game Developers Choice Awards (they lost the Game of the Year award to Portal... in which there is no shame). Despite having been pestered relentlessly all week to speak about his AI work with the title (he probably talked about it as much as if he had given a lecture on in), he was gracious enough to sit down with me on Sunday morning and chat with me about it on tape.

I want to mention a few things here first...

• I'm not a professional interviewer. I'm an AI guy just wanting to pick someone's brain.
• We were sitting in a hotel lobby so there is a bit of background noise.
• We didn't have a white board or anything else so we didn't really get "down to the metal" on any of these subjects - no code, no diagrams, etc.
• We could have gone on forever that day, but Laurie and I had to catch a shuttle to the airport.
• The one requirement his boss put on this interview was that I mention (loudly) that 2K Boston is hiring AI Programmers!

In case the above applet doesn't load or it is playing at the wrong speed, you can download the actual .mp3 file.

Here's some questions (paraphrased somewhat) that I asked and topics we discussed throughout the interview (listen to the audio for his answers and descriptions):

(0:41) How did you manage some of the issues with the potentially different factions? That is, splicers fighting splicers, the Big Daddies fighting the splicers, etc.

(5:35) Are the Splicers trying to get at the Little Sisters by default and thereby annoying the Big Daddies or is that just accidental?

(8:00) We discuss the game design mechanic of being able to choose the time and place of your "boss battle" and then actually wander around and watch the boss (Big Daddy) interact - even in combat with the Splicers.

(11:30) John talks about how the Splicers react to being cornered by the Daddies.

(12:50) What other things were you doing with the perception systems?

(19:35) We discuss how the enemies (e.g. the grenadiers) will adapt their tactics to you... both as an intelligent reaction but also to adapt the gameplay to specific difficulties in dealing with the interface and some plasmids.

(21:28) John talks about a small horror story with regard to ballistic physics and the "rifleman's rule".

(25:19) John talks about how the Daddies will use things like proximity grenades to box you in - but that how the Splicers don't even notice them.

(27:23) John talks about how the design decision surrounding whether to have the Splicers and Daddies avoid fire.

(28:13) I ask about the different tactics that the enemies use and how they select between them. Also, how they had to do workarounds to keep players from exploiting things like the pathfinding grid. He goes into detail about how this was beginning to get computationally expensive.

(34:26) What's your biggest horror story from the whole process? (John responds with an amusing anecdote about some early behavior tests.)

(37:24) What's the coolest thing that you pulled off?

(38:53) On Soren Johnson's scale from "good AI" to "fun AI", where do you think Bioshock fits?

(40:50) Do Splicers continue to spawn on a level until you leave it?

(43:00) We briefly digress into talking about the audio (which won awards).

(44:28) Not counting development time, how many times did you play through the game?

(45:24) The mandatory "2K is hiring" plug!

(49:04) John tells the story about his wonderful moment after the game is shipped. (As game developers, don't we all want to hear this?)

(Total length 50:56)

If you want to hear more of John's comments on various subjects, you can listen to the GDC AI Roundtable audio files and read my notes here. He spoke up once or twice in there.

Again, I want to thank John for taking the time with me for the interview... and being a pretty good companion throughout the week of GDC. I enjoyed it quite a bit.

Labels: 2K Boston, 2K Games, BioShock, John Abercrombie

I found an interesting tidbit this evening. It gives a look into how Will Wright wrote his first pass at the Sims' AI back in 1997. It's a scanned printout which makes it kind of annoying to read. According to Don Hopkins' description:

This is the prototype for the soul of The Sims, which Will Wright wrote on January 23, 1997.

I had just started working at the Maxis Core Technology Group on "Project X" aka Dollhouse", and Will Wright brought this code in one morning, to demonstrate his design for the motives, feedback loop and failure conditions of the simulated people. While going through old papers, I ran across this print-out that I had saved, so I scanned it and cleaned the images up, and got permission from Will to publish it.

This code is a interesting example of game design, programming and prototyping techniques. The Sims code has certainly changed a lot since Will wrote this original prototype code. For example, there is no longer any "stress" motive. And the game doesn't store motives in global variables, of course.

My hope is that this code will give you a glimpse of how Will Wright designs games, and what was going on in his head at the time!

Despite what Don asserts above, I wasn't terribly impressed since this is really something that can be written in an hour or so. I figure that is simply because it was an early pass. Of course, the Sims behavior AI is not terribly complex so it really wouldn't need to be anything much more complex than what you see here. Oh well... worth a look anyway.

Labels: code, Sims, Will Wright

One of the more intruiging lectures of the 2008 GDC was given by Soren Johnson (MobyGames info) ex- of Firaxis and now with Maxis on the "Spore" team. He was talking about how Civ 4 fit in the spectrum of game AI between two extremes... "Good AI" and "Fun AI". Here's some selections from my notes on the lecture. (Forgive the seeming lack of lucidity - I was typing like a madman!)

Also, this is a direct link to Soren's Slides (.zip) on his blog - which is where the images in this post came from (click to enlarge).

“Good” AI (Play to win)
Beat player at their own game
Essentially a human substitute

“Fun” AI (Play to lose)
Algorithms are the content
Focus on the Player’s Experience

For example, Aggro in an MMO is fun AI.

With tanks, healers, DPS (damage per second) there is a formula for handling it... Trivial AI problem to “solve” by the players.
Aggro determines who AI attacks, let enemy attack the tank, you heal the tank… DPS the mob…
Everyone knows how it works… very predictable. Almost become commoditized with agro tools. Blizzard isn’t trying to be clever – they like that it is simple.

The question with AI design is, where are you trying to fit? Across the spectrum.

• Chess is “good”
• Starcraft is more towards “good” (no real diplomacy – assumption is that they want to kill you.)
• Civ IV split the gap. Deep diplomacy but very symmetrical game design.
• Heroes of Might and Magic is more towards “fun”. Not as focused on the excellence.
• Desktop tower defense if pure "fun" AI.

Rule sets?
Good side tends towards fixed rule sets. (e.g. Chess)
Fun side tends towards evolving rule sets.

What are the best environments?
Good AI tends towards Multi-player
Fun AI tends towards Single-player

Tactics available to AI?
Good AI will do everything available.
Fun AI will do limited tactics.

Measuring performance?
Good AI has objective measurements
Fun is subjective – e.g. difficulty over performance

Turing test?
Good AI passes
Fun AI... this question is irrelevant.

The question is: “Play to win or Play to lose?”

With Civ IV, the AI does have limited options. There are a lot of options that they do not put on the table for the AI. Esp. with diplomacy. E.g. fighting a war... as a player, you can ask them for stuff if you promsie to quit war, then attack them over again. AI doesn’t do that.

Civ has:

• Both fixed and evolving design
• Symmetrical
• Single Player
• Limited Options
• Objective Testing
• Fails Turing test but it isn’t irrelevant.

Every player is different… some want things like challenge, sandbox, narrative.

For narrative, you want to aim for personality, for the AIs to maintain memory about you. It's OK for them to fall for traps. They built that into the leaders in Civ 4.

With regard to the challenge, you "want player to win or at least understand WHY they lost."

Need for difficulty levels:
Lets sandbox players off easy
Gives Challenge players a goal
Increases available tactics.

Where does cheating fit?
Completely Good AI does not.
Completely Fun AI n/a There is no concept of cheating (e.g. desktop tower defense)
In the middle… yes?

The Noble level in Civ 4 is the “even level” with regard to production modifiers, etc.

But Noble has other cheats… e.g.

• Animal/Barbarian combat bonuses
• No Unit support
• Better Unit upgrades
• No Inflation
• No War Weariness

The AI needs more help in these areas.

For example, AI does not leave cities empty like a human would… so unit support costs. Human army and AI army will never be the same size because they have to keep units in their cities… therefore cut the support costs for the AI since they need to have a larger army.

Cheats should NOT be linear… certain you want to help more or less as you progress your diff. levels.

Cheats should never feel unfair! Examples from past Civs that players hated...

Civ 1, 2
Free wonders
Gang up on human (In Civ 1: If year > 1900 and human in lead, declare war on human)

Civ 3, 4
Human-blind diplomacy (Never checks “is human?”)
Information cheats (they DO have info cheats – most of them come down to limited dev. Resources… e.g. fog of war is very expensive)

Information cheats can really backfire on you. E.g. Amphibious Assault Judo using empty port cities in Civ 3. (solved by determining random time for updating the assault target, ignore temporary data such as nearby units.)

Cheating is relative:
The Tech Trading Problem…

• AI must trade techs
• AI must trade fairly
• Human can sell techs cheaply

Only two of these can be true… So… should AI sell techs cheaply? Solution? Can AI pursue altruism?

When the AIs were trading often, it made for very even technology levels between all players rather than some groups ahead and others lagging. Everyone had everything.

Solution in Civ IV

AI can undersell by 33% but…
Tries to make up difference in gold
Only trades on random turn intervals
Uses same “Refuses Trade With” logic as with human.

Arbitrary rules e.g. “I will never trade Iron Working with you.”

What is the point of cheating?
Are we trying to…

• Write the “best” AI?
• Beat the human?
• Be fair?

Designing for the AI?
Can AI handle the options in the gameplay?
OTOH, make sure not designing just for AI. Legitimate reason for design decision. (e.g. closed borders, enforceable peace treaties)

Traditional testing fails
Automated testing helps greatly
Need hard-core fans to analyze
1.5 year closed beta, peaked at 100 users, bi-weekly patches.

They used soft-coded AI:
No AI scripts
No enums (No “Temple”)
Less brittle code
Less predictable AI is not always a good thing.

Probabilistic Reasoning - Weights to factors, values to situation

Data-driven Mods - AI was stand-alone so that it was compiled into dll. CvGameCoreDLL.dll was 100% independent of engine.

Labels: cheating AI, Civ 3, Civ 4, Civilization, GDC, GDC 2008, Soren Johnson

I'm in the process of uploading all my GDC-related things to this page. You can actually listen to my audio of the 3 AI roundtables and read my (barely comprehensible) notes that I furiously took during each. Also, it has links to the pictures that I took during the roundtables and the AI Programmers Dinner on Friday night.

On that page, I will also be posting other AI-related tidbits such as my notes from lectures such as those by Soren Johnson's (Civ 4), Damian Isla (Halo 3), and Peter Molyneux (Fable 2). Give me a few days to get it all straightened out, though.

Also, I sat down with John Abercrombie of 2k-Boston on Sunday morning and spoke with him about the AI that he did for Bioshock. That should be posted on Wednesday. Look for it over on Post-Play'em.

(Remember to tap the RSS feed to keep up with these additions and all other AI-related things.)

One final note about GDC... it's always an exhilarating week... but it sure does make my head hurt!

Labels: BioShock, Damian Isla, Fable 2, GDC, GDC 2008, Halo, Peter Molyneux, Post-Play'em, Soren Johnson

Soren Johnson, AI guru behind Civ 3 and Civ 4, posted an interesting tidbit on his "Designer Notes" blog entitled "A Farewell to Civ". He mentioned that since he is no longer with Firaxis (having moved over to the Spore team), his GDC lecture next week will really be his last hurrah in speaking about his work with the Civ series.

One thing's for sure, I will be avoiding writing about Civ 4 in Post-Play'em until after I attend this lecture. One quote caught my eye:

Essentially, I will be talking about the difference between thinking of the AI as the player’s opponent and thinking of it as simply an extension of the core game design (what one might call the difference between “good” AI and “fun” AI). There will also be a long section on AI cheating - the bane of my existence for many years - concerning which type of cheats are acceptable to players and which type are not, using Civ as an extensive case study. Further, I hope to prove that, for Civ at least, there is no such thing as - and never could be - a “fair” difficulty level where the AI is playing the same game as the human. Your mileage , of course, might vary.

I don't want to make the mistake of assuming something he did was cheating, or something I thought was cheating was actually brilliant AI work on his part!

Cheating or not, he has done some exceptional work in making Civ 3 and Civ 4 an absolute delight to play... even as an AI guy. I'm looking forward to meeting him.

Look for updates from GDC both here and in the IA News blog.

Labels: cheating AI, Civ 3, Civ 4, Civilization, GDC, Soren Johnson

Wow I love the GDC.

According to this press release, Kynogon claims they are going to have "Kynapse 5" ready to roll to show at GDC next week.

I've been skeptical of AI middleware for a while, but I believe that the industry has matured somewhat to be able to pull off a lot of things in a standardized fashion. I'm definately going to have to stop by and see what they have going on.

Labels: GDC, Kynapse 5, Kynogon

I haven't gone out of my way to look for reviews on NFL Tour. This one, however, popped the alarm on my Google Alerts.

Gaming Today Impressions Of NFL Tour (360)

Here's the quote that got me:

Unfortunately, the gameplay is just way too simple. The game is so simple to play that a monkey could easily score a touchdown on the lacking AI defense of the computer. If you want to win games in NFL Tour, just snap the ball and pass to anyone. Don’t worry… unless you are terrible, there is no such thing as an incompletion. I put in about 6 hours of gameplay and my QB accuracy is 100%.

I guess that really shows that it takes a team like the Madden/EA crowd 10... 20... 30 years of working on a franchise before they can get AI to what it needs to be (almost?). Sports AI will always be a challenge... and this title looks like they are trying to prove it.

For more scathing comments on NFL Tour, hit these...

• 1-UP
• Gamespot
• IGN

I don't do game reviews... but I think the consensus speaks loads.

Labels: adaptive AI, Madden, NFL Tour, sports games

There is a video that touches on the AI in the new Turok game. (Go to Recon/Behind The Scenes/"Awakening The Giants - Part 3") They talk about how the dinos operate with a sandbox-type AI. Also, the enemy humans operate similar to you and are capable of flanking, retreating, coop behaviors, etc. Also, they have to deal with the dinos just like you do... because the dinos will deal with them just as they deal with you.

Looks kinda fun!

Labels: Propoganda Games, Turok