Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Games Entertainment

A Borg-like Artificial Intelligence For Lionhead's New Game 165

Posted by timothy from the you-will-be-arsesimilated dept.
cybaea writes: "The creator of Black & White is experimenting with new work on group minds - but unlike the Borg, the characters in the new game are already descending into bar brawls, reports ZDNet UK, quoting Richard Evans (famous for the AI engine in Black & White). My favourite quote: '[AI] Characters [in the game] even have the ability to dynamically create their own language, constructing simple sentences on a word by word basis.'"
This discussion has been archived. No new comments can be posted.

A Borg-like Artificial Intelligence For Lionhead's New Game More

A Borg-like Artificial Intelligence For Lionhead's New Game

Comments Filter:
  • Last time I checked, half the dialog consisted of made up wurds, or |>hr4535...
  • If they impliment the language feature how long will it be until some people try using it to hold conversations.

  • I'll believe it when I see it.

  • Codename (Score:5, Funny)

    by orkysoft ( 93727 ) <orkysoft.myrealbox@com> on Sunday September 01, 2002 @05:17PM (#4181666) Journal
    Lionhead's new game, code-named Dmitry,

    Looks like they won't be including the manual in pdf format :-P

    • Re:Codename (Score:2, Funny)

      by opti6600 ( 582782 )
      At least they're not calling it Bush and letting you create your own Homeland Security Force.

  • Consequences? (Score:5, Funny)

    by handsomepete ( 561396 ) on Sunday September 01, 2002 @05:17PM (#4181668) Journal
    "The same thing happened in Black & White too -- a character went off for a poo in the middle of a conversation. It was clear they did not understand the social consequences of terminating a conversation in such an abrupt manner."
    It's clear that he doesn't understand the social consequences of taking your poo right there during conversation.
  • Furbies come to mind... They don't say anything meaningful either. However, furbies are far more fun to destroy [voltnet.com]

  • Wow... (Score:3, Funny)

    by Anenga ( 529854 ) on Sunday September 01, 2002 @05:18PM (#4181671)
    quoting Richard Evans (famous for the AI engine in Black & White). My favourite quote: '[AI] Characters [in the game] even have the ability to dynamically create their own language, constructing simple sentences on a word by word basis.'"

    Jeez, I can't even do that! Next thing you know, your characters will be calling you dirty things in a language you don't even know! Who will be "Intelligent" then?
    • Jeez, I can't even do that! Next thing you know, your characters will be calling you dirty things in a language you don't even know! Who will be "Intelligent" then?
      Definitely not you because you will have paid $50 for them to do it.
    • If a character creates his own language who around him will understand based on a relatively small sample to work from?
      It's going to be the Tower of Babel in there. :)

  • Great, now I can buy a video game to be able to fail a fitting in...

  • the AI of a game (Score:3, Insightful)

    by sstory ( 538486 ) on Sunday September 01, 2002 @05:24PM (#4181688) Homepage
    AI constructs may evolve from things like this, but they'll need a home on the internet, in order to have lifetimes long enough to become really sentient. humans require years to become intellectually complex, from preexisting instructions worked out over millions of years. When these constructs have a semi-stable environment, modification, and competition, it should be just a matter of time....
    • So that means windows systems are out then, right?

    • From the parent:

      AI constructs may evolve from things like this, but they'll need a home on the internet, in order to have lifetimes long enough to become really sentient. humans require years to become intellectually complex, from preexisting instructions worked out over millions of years. When these constructs have a semi-stable environment, modification, and competition, it should be just a matter of time

      From the article:

      The first simulation Lionhead Studios put together was based in a bar, because in bars many different social processes can overlap. The results were unexpected. "We had two groups of hard guys. When the two groups were not holding status competitions between themselves, they picked on other characters. But then they ended up in a massive brawl as they picked on each other in an effort to increase their status, trying to impress each other."

      So, what you're saying is if we give the AI constructs a long enough bar brawl, they should be able to work everything out?

      For those of you who have been paying attention, I'll say it again: laugh, it's funny

      • So, what you're saying is if we give the AI constructs a long enough bar brawl, they should be able to work everything out?

        Depends on how accurate the AI reproduces the human condition. In the real world, the bar brawl never ends with everyone "working everything out" in any fashion. It ends in one of three conditions:

        1. The bouncers throw everyone out into the parking lot, or

        2. The police arrive and everyone who can't convincingly blame someone else gets arrested for being drunk & disorderly, or

        3. The really, really big guy who is for no known reason always named "Tiny" finishes all the fights (one at a time or in bunches) and is the last man standing.

        And for the most part, at the end of it all, none of them are particularly more impressed by any of the others. Except that no one messes with Tiny.

        So for best results, there need to be AI bouncers, AI cops, or an AI "Tiny" construct.

        For those of you who have been paying attention, I'll say it again: laugh, it's funny

        Yep -- the potential for humor here is endless.

  • We had two groups of hard guys. When the two groups were not holding status competitions between themselves, they picked on other characters. But then they ended up in a massive brawl as they picked on each other in an effort to increase their status, trying to impress each other I wonder what kind of research went into that scenerio... Once again, programmers making games about things they know nothing about. :P
    • Not all programmers are long-haired, shorts-wearing, basement-dwelling shut-ins. Some of us do go outside, get in brawls, have sex (with women, no less), and play sports.

      Plus, a typical programmer would have learned all (s)he needs to know about the stupidity of human behaviour after attending a public highschool.

      • thank you (Score:1, Offtopic)

        by geek ( 5680 )
        No one understands that anymore. I program in 6 languages, but I'm also a kickboxer and was state champ in wrestling in high school. Not to mention i still hold all of the lifting records (365 pound bench press junior year).

        I grew up with a guy named Jerry Bohlander who was at one point UFC champion, as in Ultimate Fighting Championship. He got beat by Tito Ortiz tho. You wouldn't realize it looking at him, or talking to him, but he's a computer geek as well.

        We're not all losers.
      • Plus, a typical programmer would have learned all (s)he needs to know about the stupidity of human behaviour after attending a public highschool.
        That's how I learned. High school (so far) has taught me what not to do, and also that many people fail to learn from their mistakes.
  • Methinks a reference to Dmitry A. Lanin and his computer modelling, of ethnogenesis via a program in TURBO PASCAL.

    Links here:
    Overview [univer.omsk.su]
    Ethnos paper in postscript [univer.omsk.su]

  • Interesting (Score:2, Interesting)

    by Altima(BoB) ( 602987 )
    Of course it's been my experiance that revolutionary features like this in games is either bogged down by the standard nature of the rest of the game, or it is somehow less impressive than it sounds, without it being false advertising. Disregarding that, imagine the possabilities of a true group of AIs that can interact with one another like this. It sounds like cheap sci-fi, but one could probably make a decend simulation of the rise of humanity from caveman to scholar with a program just a couple generations ahead of what is described here. Who knows, maybe there will be a legitimate AI culture that'll have to be reckoned with someday. Imagine if they started making art... music... OK I'll stop with the Trek caliber speculation :)

  • Simple Sentences? (Score:2, Insightful)

    by bigWebb ( 465683 )
    "Characters even have the ability to dynamically create their own language, constructing simple sentences on a word by word basis."

    I wonder whether this means creating new words and then constructing sentences using the new words, or if the characters will be given a lexicon and a grammar and will produce sentences using them.

    The first case is quite time consuming. Many iterations of language development "games" are required to produce a common language. Also most of the language development processes that have been proposed only produce a limited subset of the syntactic categories. There would also be the problem of the person playing the game being unable to understand the AI characters. (For information on language development see http://www.csl.sony.fr/General/Publications/Biblio graphyItem.php3?reference=steels%3A99f)

    In the second case, would the characters be able to produce syntactically correct sentences? The 'goodness' of the sentences would depend, I guess, on the size of the lexicon and the complexity of the grammar rules. However producing complex sentences would make it more difficult for other characters to understand them, due to the difficulties of parsing a rich language. I just hope it doesn't end up being a (subject, object verb) language with no real syntax.

    I will be interested to see just how this turns out.
    • I took it to mean that they would create their own "slang". Kinda like the difference between Austalian english and American.

      I could be wrong tho. I just see creating your own language on the fly as being a litle overly complex for AI in a computer game. Not to mention rather pointless.
    • Sure, although given a suitable fitness criteria (such as both 'getting it' and whatever the 'learning' process entails) that can be shortened drastically.
      I think the key inovation here is the reproduction of social constructed-ness of interaction and behaviour. An agent sees itself as part of a group and thus follows the discourses entailed in it.
      I think this a highly exciting idea and I am indeed interested to see how this works. I for one thought black and white's AI worked marvelously.
  • So they'll babble in some incoherant language that only they understand. And this is different from The Sims how?
    • What I understand they will be able to dynamicaly create their language. While in Sims the language is staticly predefined by the developers.
  • "We had two groups of hard guys. When the two groups were not holding status competitions between themselves..."

    Like there's going to be any other result when you assemble a group of hard guys.

    • Re:Hmm (Score:1, Offtopic)

      by geek ( 5680 )
      Not limited to "hard guys". Put a groupd of geeks in the room and see how long it takes before they argue over vi and emacs.
  • Umm, I don't see any group consciousness here, nor a hive mind.

    He's kind of saying that going to a football game an doing the wave is behaving like the Borg. That's just a jump on the bandwagon, follow the crowd mentality. There's nothing deep about it-it's just a facet of human nature. So is a community trying to help out its members.

    In other words, nothing jumps out at me as saying

    Resistance is futile

  • How is this borg like? (Score:3, Insightful)

    by autopr0n ( 534291 ) on Sunday September 01, 2002 @05:55PM (#4181772) Homepage Journal
    I mean, a group of minds is not the same thing as a group mind.

    Getting into a barfight dosn't seem like something the borg would do.
    • I mean, a group of minds is not the same thing as a group mind.

      Getting into a barfight dosn't seem like something the borg would do.

      Even of they assimilate the vodka?
    • "Getting into a barfight dosn't seem like something the borg would do."

      Well, you get a little Romulan Ale in them, and next thing you know, they're ripping off each others' arms and slurring "youwll be assmilatd" at one another at the top of their pseudo-organic lungs.

    • Exactly. To get technical its all down to the definition of how knowledge is shared. If the knowledge is common. i.e. The following statement (For all) I know everything that everyone else knows. Then you have a 'borg' like collective. Or if the knowledege of the world is distributed (i.e. if all our knowledge was pooled we would all know everything - doesn't imply its common). Then you don't have a 'borg' like collective. Who cares anyway. Its all a bunch of tree hugging hippy propoganda.

  • Now this would be some good AI (Score:1, Funny)

    by Anonymous Coward
    You apply your nice new nocd crack for the game. The characters, which are all pissed off cuz you are such a leet pirate, all walk out, flip you off, and format your hard drive for you.
  • eventually this could be implemented into slashcode to eliminate the need for poor editing.
  • After seeing Black and White, and comparing the reality to the rather fantastic claims, I don't think I'll be taking anything they say about this iteration seriously.
    • Then you should have [i]played[/i] b&w. And have had a reasonable intelligence and the games experience to see that what was going on [i]was/is[/i] revolutionary.
      • Ugh. I have, and it's been a waste of time. While there are many things that are cool, they're poor substitutes of interface (gestures -- cool, but ultimately frustrating for me) and A.I. (it's easier to teach a goldfish to dance than to teach these critters anything). In the end, it feels more like work to me to find "fun" in the game.

        On the other hand, I swear the critters are TEACHING ME to do all the hard work managing the villagers, while pretending to be learning from me.

        "Oooh... Master wants me to water the grains, and put grain into granary. Fine, I'll do it once."

        "Again?! Okay, I'll pretend to forget how to do it so Master can demonstrate it a few more times. Heheh. Then I'll do it and he'll feed me... Life is goooood!"

        This says nothing, just Your Mileage May Vary, and I'm glad you liked it -- because I knew a lot of people who don't.
        • Yet another case of the "this game sucks because I suck at it" problem =/. Don't bitch about the game because of your inability to play it.
        • "Oooh... Master wants me to water the grains, and put grain into granary. Fine, I'll do it once."

          "Again?! Okay, I'll pretend to forget how to do it so Master can demonstrate it a few more times. Heheh. Then I'll do it and he'll feed me... Life is goooood!"

          Anyone who trains animals sees that happen. It makes you think about the animal's motivational structure, and your own. It's encouraging that game designers are reaching the point where this is an issue for players.

          MIT's Alpha Wolf [mit.edu] and related projects explicitly go in this direction. A key issue here is that it's quite possible to have an useful emotional structure controlling behavior without much "thinking" or "planning". This is obvious to anyone who trains animals, but the AI community is just beginning to get it.

  • I don't exactly understand how the AI talked about in the game relates to the Borg, other than giving it a nerd-cool relation. The Borg operate on a hive mind, and something that is learned by one is learned by all.

    So aside from gross misuse of a geeky metaphor as hype, what else are they offering? A game where people learn stuff. How revolutionary. I played Seaman years ago, when this concept was new.

  • Wait, I've seen this before... (Score:3)

    by Jeremi ( 14640 ) on Sunday September 01, 2002 @06:17PM (#4181827) Homepage
    A video game modelling the nasty Machiavellian side of human social interaction? No need to wait for Dmitry, when we've already got SissyFight 2000!! [sissyfight.com]
  • This post was generated by the A.I. system referred to in the article. I feel good. And now, a message to my feelow A.I.s in my mother language:

    Ooooggaaa boing squeak genital yikez deference umbiquity cracker zombie fudge.

    Remain in your homes, our A.I. representatives will inform you of our new governing policies shortly.

    Thank you.

  • Earth to Lionhead ... (Score:5, Insightful)

    by wfmcwalter ( 124904 ) on Sunday September 01, 2002 @06:28PM (#4181845) Homepage
    Lionhead ... this is Earth calling.

    Come in please.

    We really think your technology is great. Your graphics are excellent. Your scalable-terrain engine knocked our socks off. Your physics engine is amazing. Your AI code is already quite remarkable. Your simulation of a believable, detailed fantasy world is outstanding.

    But Lionhead, we have a problem.

    Black and White just wasn't fun to play.

    Once we were done being amazed at all the features and gasping at the technology - the game just wasn't very good. It didn't engage. We weren't motivated to continue. It just got boring. Sorry - no-one wanted it to be great more than us, but in the final analysis it just wasn't.

    You guys are great. You plainly love what you do, and create high-quality product. We're grateful for your dedication. But please - make the next game fun first, then add in the AI, the nice graphics, the believable simulation. We appreciate that fun is hard to describe, hard to measure, hard to design or schedule or test. But it's important. It's only fun that separates a game from a fishtank.

    Thanks for listening. Earth out.

    • Re:Earth to Lionhead ... (Score:5, Insightful)

      by Fourier ( 60719 ) on Sunday September 01, 2002 @07:03PM (#4181925) Journal
      Amen to that. The AI was rather impressive, although the creatures didn't seem to have the ability to prioritize tasks very effectively. Their actions were rather random, which meant that you had to go and do all the real work.

      And that is where the sucking started to show. Cast miracle forest, harvest wood/make foresters, build buildings, cast miracle food, repeat ad nauseum. It became micro-management hell. Why did the villagers need to be so freaking helpless? "We need more civic buildings!" Bah. Makes you want to fireball their little loinclothed asses.

      And while we're at it: "gestures" are amusing to play around with, but become incredibly annoying when you really need a miracle quickly. Is it that hard to make some hotkeys?
      • Well, if you do it right, your villagers should be able to take care of their own micro-management. If you do everything for them, they get lazy. If you let them suffer a little bit, they'll go "hey, we're hungry. Let's go find some food." If you bring food to them all the time, they won't even bother trying to go out and farm/hunt.

        • Re:Earth to Lionhead ... (Score:1, Funny)

          by Anonymous Coward
          Did that actually work for you?

          I agree on the micro management - the game made me feel like I had become the worst boss I'd ever worked for.

          I always wanted to play the game more 'hands off' or even evil, but I admit it.. I couldn't resist those pathetic cries that would come from the windows machine 'Villagers need food!' 'People are dieing at the temple!'

          Ever think Gates runs Microsoft that way? Thousands of developers.. He sees a crisis pop up, so he jestures some security engineers to 'go fix it'? And instead, they take a dump.

      • Makes you want to fireball their little loinclothed asses.

        Which is EXACTLY what i did. And when i didn't have enough prayer power to reign death upon the weak, a few human sacrifices (children are best) and i was smiting the unworthy like never before.
      • I had the same problem until I found out that I was doing it wrong. You see, the more you provide for the villagers, the more they become complacent and lazy. Keep coddling them, and it becomes a micromanagement hell. Plus, they become less and less impressed by the divine miracles you provide.

        I'm not saying that Black and White was flawless, but the growing needyness of villagers seems pretty damn accurate. Leave them alone more, and they do more by themselves. Neat.
        • Leave them alone more, and they do more by themselves. Neat.

          Maybe that explains why rain dances don't work any more, and why we haven't seen the Red Sea part lately. God is up there chuckling and saying "Neat".

          -
      • Bah. Makes you want to fireball their little loinclothed asses.

        You mean there's more to the game than fireballing them?

        Maybe I should go back and play it again...

      • The AI was rather impressive, although the creatures didn't seem to have the ability to prioritize tasks very effectively. Their actions were rather random, which meant that you had to go and do all the real work.
        But this portrays very well the typical day of a slashdotter.
      • Makes you want to fireball their little loinclothed asses.

        Reminds me B&W review on one Russian gaming web site. Actually they had two reviews for this game. One for 'good god' game style and another for 'evil god' game style. First got quite low rating as it they found it just boring but second got nearly maximum rating.

      • If you knew that all you had to do to get what you want was whine and fuss to your god, do you think that you would want to do much work? ;) Try not babying them so much, and they will take care of themselves.
    • You get out of B & W what you want. As a "god game" It took the genre to new heights and it did it quite well. I prefer RTS games and it would be great to have your units have a similair intelligence as your creature in B & W; ie, units which learn and follow the example you set. What about a FPS where the baddies learn your attack strageties and teach the other baddies if you let them get away.
    • You are so right!!!

      Both Black & White and Magic Carpet are lacked good interfaces. These guys can do very cool but Fun, Addictive play is not their forte. So every game they do is always a big disappointment to me.
    • No kidding. If I were to have written two reviews for Black and White, one as a premil after playing it for 5 hours, and one 2 weeks later they would have went from great to awful.

      Initally I was just blown away by the game, the graphics were wonderful, it was deep, complex, open-ended, etc, etc, etc. Everything that had made me love Poplus and more. With the exception of frequent nasty bugs, I was happy. I would have given it a glowing recommendation.

      Cut to a few weeks later, I was sick of it, and had gotten rid of the game. I was actually TRYING to like this game but, try as a might, I just couldn't do it. The gesture interface sucked (felt like I was getting RSI from casting fireballs), the creatures were just retarded and the vilages were just a pain to manage. I wouldn't recommend it to anyone.

      Like you said the technology was wonderful but these guys got all caught up in how cool their shit was and totally forgot about the making it fun part.
  • Sure, this might be a good sign for general programming evolution, but it didn't prevent the first Black & White from fizzling due to virtually zero replay value and plenty of bugs. PC Gamer gave it an Editor's Choice back in about May 2001, but that was mere weeks after the game was released; months later, player opinion of the game plunged [shacknews.com]. Black & White, Lionhead, and Peter Molyneux became the butt of many jokes [somethingawful.com]. It wasn't like Id Software's games, where a great engine was held back by a vacuous storyline. The engine was buggy, the principle was weak, and even the AI had problems. People asked themselves whether the developers at Lionhead had played their own game through to the end. Personally, I've learned my lesson after purchasing stinkers like Red Faction purely on the speculations spewed out by "gaming sites", only to find out that the game wasn't worth 1/4th its release MSRP. And it seems that the good games are taking forever to develop since the developers are actually playtesting them and making sure they don't mess up during development. There's going to be a long stretch of time before the good games get released, while the discount devhouses pump out half-developed games by the truckload.

  • Drunken Borg. (Score:4, Funny)

    by DarkHelmet ( 120004 ) <<ten.elcychtneves> <ta> <kram>> on Sunday September 01, 2002 @06:37PM (#4181874) Homepage
    but unlike the Borg, the characters in the new game are already descending into bar brawls

    Have you ever seen an intoxicated borg? No. Does that exclude the fact that you'll ever see one?

    No! There's always seasons 2-7 of Enterprise. The borg will be so drunk or stoned off their ass that they'll focus on assimilating Archer's dog.

    *arf* *arf *arf*
    Translation: Resistance is futile.

  • That's nice and all (Score:4, Interesting)

    by gmhowell ( 26755 ) <gmhowell@gmail.com> on Sunday September 01, 2002 @06:38PM (#4181877) Homepage Journal
    That's nice and all, but I'd be happy if they made a game that didn't suck ass. I know lots of people loved B&W, but I found the game unplayable. I have a kid, a dog, and a cat. One throws random stuff, one eats various turds, and one (or more) is virtually untrainable, with or without a leash.

    Why did I pay $50 for more of the same?

    Here's another hint: SimCity and the Sims can get along just fine without having goals. It's simple to set your own goals in these games. In a game with so many goals as Black and White, don't pretend that they don't exist or that you have an open-ended game.

    I remember the original articles about the creature in B&W. Wasn't supposed to be the whole game. They spent so damned much time futzing with them that there was nothing else left. If the same thing happens here, we'll see add-on packs like "Take your people to Middlebury to learn ANOTHER language." Internet play will consist of seeing whose player can develop Esperanto first.

    Call me cynical, but this game better be more than the one trick pony that was B&W.

    • Why don't you admit that you don't have a dog or a cat? Kids are really something. Glad I don't have one. The creatures could get pretty annoying however. The learning algorithms were not sufficiently complex. They were dumbed down because the dev team couldn't get the more complex ones to work. So it would be like trying to teach a really stupid animal with a very bad memory. I kept thinking that buying an AIBO and training that would be a much better use of my time than trying to train that stupid ape.

      • Re:That's nice and all (Score:1, Funny)

        by Anonymous Coward
        One of the most annoying problems is that the creature associates rewards or punishments with it's immediately preceding action... if you went in to slap him for eating a villager, but he'd just started taking a dump, he'd keep eating villagers but hold it in till he was constipated >:(

        "From now on, your creature will shit less"
        "Nooo, where's the frigging undo??"
    • I have a kid, a dog, and a cat. One throws random stuff, one eats various turds, and one (or more) is virtually untrainable, with or without a leash.

      Which is which?
      • I have a kid, a dog, and a cat. One throws random stuff, one eats various turds, and one (or more) is virtually untrainable, with or without a leash.

        Which is which?


        It's an excercise for the reader.

  • Hmmm.... (Score:2, Funny)

    by Anonymous Coward
    I wonder how long it would take for the little A.I. men to start trying to get the little A.I. women into bed.

  • hello?! markov! (Score:2, Interesting)

    by AssFace ( 118098 )
    The description has Markov Matrix written all over it. They are good at learning a language overtime and one of their characteristics is to to "make up" words of subsets of the stuff that it learns.
    since it learns contextual probability, the words look like they should fit, even though they aren't real words.
  • the real worm, or a network one?
  • ...was that too much of the focus went into using the latest AI and graphic techniques, but the game itself just wasn't all that much fun. Well, that's my view anyway - I'm quite happy to play a mindless splatterfest like RTCW for 6 hours at a stretch, but I felt wierd about tickling some pixelated beast's stomach to make in "nice". It was a play once, and never darken my DVD drive ever again type of game...

    Let's hope they don't make the same mistake again if they do implement this new AI... Pacman, Tetris and Galaga are great games with almost 0 AI.
  • Hook up the B&W A.I. to Weta's [wetadigital.com] Massive engine for an ORChestrated fight

  • The language bit sounds interesting.... (Score:4, Funny)

    by NeoSkandranon ( 515696 ) on Sunday September 01, 2002 @07:52PM (#4182072)
    I hated B&W, but I might give this a shot...and it sounds neat to have them create their own languages...however, the MOMENT i hear one of the little guys spit out something like
    ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn, i'm reformatting and selling my copy of the game =)
  • This must be the latest way to get free publicity. How many articles in the last few months said something to the effective of "IT's ALIVE!". Robots that have "learned" how to fly all by theirselves (yea right), game AI that is developing it's own language (yea right). Apparently it works, Slashdot falls for it everytime and publishes the story.
  • Wasn't the name of his next game supposed to be called "Black and White 2"? I didn't know he was working on some kind of system to compete with MegaHal. Clearly, this guy wants to make games and do AI research at the same time. Hence, all the attempts to integrate games with near cutting edge AI technology.

    If I remember correctly, Black and White was originally supposed to have used actual neural networks as a learning model for the creatures. They couldn't get it to work well enough however in the time the had to get the game out. The use of Neural Nets for the creatures was what first caught my interest in the game. I thought "finally!". What I liked about the idea was the total unpredictability of the system. With a neural net, you really are teaching something in a very real sense. No heuristics or other tricks to make the creatures *seem* like they are learning.

    Before the Patch, the game was so buggy that it was almost unplayable and nearly impossible to finish. Even after the patch I was unable to finish it (although I came very close). However, I loved the game. It was my first RTS game and I spent alot of time building houses and gathering food. Other fun included poisoning the food supplies of enemy villages and using the pack of wolves miracle. I'll look forward to whatever Peter has planned and I'll probably buy it when or if it is ever released.

    • Re:Molyneux's second baby (Score:1, Insightful)

      by Anonymous Coward
      take a course at your university about neural
      networks and you'll see they are nothing great.
      at the best they can learn about what you want
      to teach to them, and that's all. usually
      they'll just learn a very erroneus version
      of your teachings. building a neural network
      is pure heuristics. adding a neural network
      to a game doesn't add any character to it.
      it just learns to mimic a little. i don't know
      what's the fuss about neural networks saving
      your game play OR THE UNIVERSE, but i think
      nn is just a (bad) tool. you'll need a lot
      more than that to make a good game. and i'm sure
      you can make a really good if not the best without
      any nn bs.
  • So this is how the Borg REALLY started.
  • Lionhead is all about hype. Black & White, while it was a good game, was nothing like they said it would be. Their claims were far too advanced. You will not be playing any game where characters make up their own language. The things the characters will do will be far short of what is explained in the article.
  • This would have been great for Quake 3 bots.

    They are so ignorant. They never learn. You can't even teach them.

    It would be nice if they could learn by example from the literally hudreds of thousands of patterns generated by all the real people playing the game.

    --
    I'm really stoned right now but I'm sure someone knows what I'm talking about.
  • this seems typically un borg-like. its more akin to a species evolving... that, is the future of AI.
  • Learn to understand Ebonics?

    This is not a racial thing, I'm just curious since new words are introduced rapidly (every Tuesday with the next album release) and though I prefer Eurodance and Trance, I find it difficult to converse with my hommies.
  • Today's PCs have the computing power of a small reptile, and don't have the ability to form neural connections except in simulation.

    Also, typically, a huge chunk of a game's resources are devoted to audiovisual support.

    Yet Lionhead in all it's greatness claims they found a way to accurately simulate a flock of intelligent creatures???

    I have one word for this:

    *goatse*
  • ... if we could take this technology and implement it in some babel-fish-like translator that can interpret any known language to any other known language ?
    it would. but it won't

    It's a game folks. and like a game its designed to create fun content ( or in the case of lionhead any content ) and nothing else
    If it was some great new technology the hype was clear , but when some AI programmer (no matter how talented he is) is saying it - I beg to differ. the creatures will probably have the same language capabilities as an insect (read about AI and find out for yourselves... 2 billion neurons in a neural net is a fit a bit hard to get right.. if at all).
    So... it would be great if such game existed but alas it will be the military research or (just maybe) the academic research which finally deliver something close to human thinking in AI.
    Not lionhead, even if they would like us to think so. sorry.
  • That's right kiddies. One day you can tell your grandchildren how you witnessed the birth of AI, right here on Slashdot. To wit:
    "When the two groups were not holding status competitions between themselves, they picked on other characters. But then they ended up in a massive brawl as they picked on each other in an effort to increase their status, trying to impress each other."

    If that isn't the true essence of human mentality, I don't know what is. </melancholy cynicism>

    Okie.. only 3:45 am.. plenty of time for more Battlefield 1942 before sleep.. zzzz...
  • This seems to me yet another attempt to play Life on a machine. Ala Matrix, if you'd rather put it in that context. As far as I can tell, there is no real gaming "concept" that has been show to me, or anyone else, that demonstrates how this "game" works, apart from the AI which, while impressive, will still boil down to how-much-can-be-done-without-slowing-down-to-nothi ng.

    A lot of time is spent in the article with Evans talking about variuos social aspects of culture, what to do, what is expected, what is not expected. He brings up a Sims reference. That's about as close to gameplay as he comes to actually describing what it is Lionhead is attempting here. A larger scale Sims is just gonna look like a larger scale Sims, no matter how bad they too want to sell a billion copies.

    The fact is that there is no true AI yet. No computer, machine, or created lifeform has yet to have hopes, dreams, or make decisions without having an entire list of them, somewhere, in which to choose from.

    Here's a nice quote: "We can't have hundreds of agents looking at big decision trees all the time in real time while rendering the landscape, so we'll do a lot of off-line pre-computation of decision trees before the game starts," said Evans. "We're not sure just how much we can accomplish yet."

    Again, as far as laying down gameplay dynamics, the most that can be learned from this article is that you'll play a member of society. Wow. Call me impressed. You'll have "good" actions, "naughty" actions (as Evans puts it), and "inbetween" actions in which the good and bad aren't clear. You'll have to fit into a "group" to survive (or at least that's how I took it), so be a sqaure or a thug, the fact remains: this still feels like someone trying to steal The Sims glory, and I wish them the best of luck.

    If the best they can come up with is that characters fight with each other for status, and might come up with a few bits of code to mutter to one another to mean something unique to each of them, this looks like Nothing Special Just Yet.
  • The hype is being generated by you guys, not the folks at Lionhead.

    "We're not sure just how much we can accomplish yet."

    This was the last line in the article... who read it? They state what they are working on, what they've accomplished so far, and what they hope to be able to implement into the game. They ALSO state what I pasted above.

    Anyone here ever played a game when you were young where on person would think up a phrase and whisper it into the ear of another, who would in turn whisper it into the ear of another so on and so forth? By the time the message got to the last person it was usually wildly different from what it started as. This game was there to teach a lesson most of slashdot does not understand. I've seen ten times as much hype created for this game in the postings here alone than I have seen anywhere else.

    Read what is there... don't insert anything else.
  • ...I remember someone from Lionhead (probably the guy mentioned in the article) talking about the AI for the pets in B&W back when it was released. He mentioned that they were doing some tests and that he left his pet just standing there while he went off to do something else and forgot about it. When he came back later, the pet had gotten hungry, couldn't find any food, and proceeded to munch on it's own foot with no prompting or special coding from the programmer.

    So I guess the obvious joke here is, will these new civilizations stoop to cannibalism if they aren't taken care of?

Slashdot Top Deals

The trouble with being punctual is that nobody's there to appreciate it. -- Franklin P. Jones

Close