Why can't we ever have decent AI?

[Matri]

There was a pong table game which burned you for letting a goal in. It was made by Germans, go figure. Apply that, to say, Unreal.

I'd rather not. The lightning gun's gonna... yeah...

And I'd like to point out that even if the pain is non-lethal and temporary, being on the business end of a chaingun in motion is not going to be very pleasant at all. Not even in the name of fun.

 

But a forcefeedback suit, for when we migrate to a full VR environment... I wouldn't mind a set.

[KoMik]

About humane AI. I haven't met games where I could (once in a while) shout to mobs something like: "Freeze", "Surrender", "Drop your weapon", "Join me", "Hitler is not the fühler", "Your mama is an awful and smelly w*ore" etc. Then mobs should react (you know how) or shout back and try to make gamer react. Intercourse.

 

When AI is wounded (human shaper), it hardly ask mercy. Morale and propaganda is also a big factor in wars/conflicts.

 

Does these belong to games? I think no, but I'm not sure.

[Krabjuice]

Overwatch don't beg, they're not programed to.

[Crazy Gringo]

About humane AI. I haven't met games where I could (once in a while) shout to mobs something like: "Freeze", "Surrender", "Drop your weapon", "Join me", "Hitler is not the fühler", "Your mama is an awful and smelly w*ore" etc. Then mobs should react (you know how) or shout back and try to make gamer react. Intercourse.

 

When AI is wounded (human shaper), it hardly ask mercy. Morale and propaganda is also a big factor in wars/conflicts.

 

Does these belong to games? I think no, but I'm not sure.

 

AI human interaction would be nice....The most I have seen it is in Vietcon where there are group commands that seems to 'overwrite' the AI aka. Free Will of your group members who some times get a bit too gungho and wants to go storming after the enemy with out care of their personal safety.

 

Also as to how your persona/alter ego in the game act in his/her/it's surrounding would be a nice feature....Running through a crowded populated area with weapons blazing should make the AI controlled persons aka. mobs behave in three ways:

 

1. Cower in fear

2. Run screaming away

3. Draw their own weapons and start blasting away at you or pommel you to death with their fists/meele weapons.

 

I have only seen two games where the AI has behaved like this....It was in Postal 2 and Mafia. Even if the AI were pretty screwy some times in Postal 2 like when you have beaten a mob in submission/cowering in fear and you pi** on it will freak out, get all brave and try to kill you.

 

I am getting the Godfather game this coming week so let see how the humane AI behaves in the there. (Plots some mass hit-and-run with speeding vechicle Mafia-style).

[Thorondor]

"Why can't we ever have decent AI?"

 

How about half-decent ? Well, I know what you mean...

 

I do think, for starters, that Olav - much as the rest of us - has the right to feel cheated out of, really, basic improvement.

 

There are many examples of such failings, but one that regrettably comes to mind, for instance, is UFO: Aftershock. I can recall roughly pointing out the more glaring loopholes in UFO: AM's AI's behaviour way back in November 2003 (thus, only a month or so after the release of the first title) and, for the most part, the same problems still plague the sequel two years its senior.

 

What could we fairly demand of ALTAR Games in a next iteration in 2007 ? Plainly more than what we've gotten in the previous product cycle's "leap" in this respect.

 

Having read the previous posts I find that, even if this thread isn't going to solve any of the real problems standing on the way to success, it still works as a repository of experiences and individual perceptions that can work towards a better understanding of the fuzzy purported goal - an Artificial Intelligence that 'feels' right.

 

Personally, my desire to see games with an improved AI component has little to do with any inability of current titles to present adequate difficulty or challenge (which can always be provided for in one way or another) but has instead to do with the supplying of sufficient believability and fun.

 

Fun being very much at the core of my reasoning when making the case for investment in better AI. According to studies *adopts professoral stance * to our brain novelty equals fun.

 

Finally mastering a concept allows the overcoming of a challenge, which rewards the process of learning with a pleasant result - winning. Winning again by employing the same methods is then increasingly less fun until, eventually, it defaults to boredom (that's the part when you lose sight of the game's box underneath all that dust ).

 

A few things emerge from this

- logic drives learning

- poor logic is no fun

- losing is the part of learning that leads to winning

- losing without fairness is no fun

- winning is the part of learning that leads to fun

- winning without learning is no fun

 

 

Naturally, it follows that when we are faced with a predictable foe that cheats to mask stupidity, boredom and frustration can never be very far away.

 

Predictability is still desireable, but only to the extent where it may lead to learning by deductive thinking.

 

The same is true of the poor logic bit. Working both ways, when either you or the AI foe noticeably underperform without discernible logical underpinning, no fun is on sight:

 

> an enemy that acts with little or no regard for self-preservation, is displaying poor logic. If, however, it is an irrational being or non-sentient implement, then valid logic is possibly at work and has no detrimental effect on enjoyment until proof to the contrary is made evident.

 

> an enemy that acts without discernible purpose is displaying poor logic (wanders back and forth without target-fulfilling destination, for example). If, however, the behaviour derives from panic or lack of sanity then valid logic is possibly at work (grounded atypical behaviour) and has no detrimental effect on enjoyment until proof to the contrary is made evident.

 

 

The latent irony in all this is that the future satisfaction of our demands lies largely in a developer's ability to keep poor logic from becoming obvious even when it really is.

 

There's progress for you...

 

::

 

Ultimately, an artificial intelligence that acts believably and with a discernible purpose is, IMHO, what we're looking for at the finishing line. If that means that some behind-the-scenes tinkering has to take place to maintain the illusion, then so be it.

[Matri]

I agree. AI that presents a challenge is good, but being unbeatable even on Easy isn't desirable either. The problem I can see is that it's not easy finding a "sweet spot" between too stupid and too smart.

 

Still, you have to admit AI's evolved from the days of Wolfenstein 3D and Doom, where they just point them at the player and shuffle forward, oblivious to walls and obstacles.

 

EDIT: Something just occured to me: Pathfinding.

 

Having an AI with the intellect of Einstein means nothing if "Einstein" can't find his way to the other end of a 3-feet corridor. The enemies in F.E.A.R. can't be called smart if they had to run to the exit behind you to get to a safe location instead of the door behind them. And someone's comment earlier about jumping off a bridge to certain death to escape a grenade instead of backpedalling.

[Thorondor]

@Matri:

 

"Still, you have to admit AI's evolved from the days of Wolfenstein 3D and Doom, where they just point them at the player and shuffle forward, oblivious to walls and obstacles."

 

Hey! Don't diss Wolfenstein 3D... (I love that game !)

 

AI has evolved, yes (GalCiv II is a _great_ recent example of this, with a scalable AI to boot).

 

The problem is "every game is a game", so to speak. We often complain (and fairly so) that game X got this and this right Y years ago, so we should be shown the courtesy of getting an, at least, as competent performance on those areas - which, as we know, doesn't always happen in subsequent titles of a franchise, let alone a genre. You spoke of pathfinding, well, I rest my case right there...

 

When push comes to shove, I'm forced to admit it doesn't feel like it evolved half as much as I honestly expected it to a few years back.

 

For instance, I see you've mentioned a game called 'Z' by Bitmap Brothers on a previous post (a have the game's box at hand's reach this very moment, BTW ). When it came out it offered an interesting, fresh, twist to the genre and a fun challenge with its territory capture mechanics. It was also fun, as I remember:

 

- to blow buildings (and assorted terrain) up and watch the debris rise "towards" the player, the little fragments causing damage upon falling

- to pick off tank drivers to take over the enemy's armor

- to see those two grunts slacking off and drinking beer while general Zed set things ablaze

- to watch your troops playing cards or fishing in ponds while idle

- to blast both the enemy and the little native critters running around that populated the world (I'd have a Tough make a splash by firing at crocs ) and see them picking themselves up

 

Bottom line is it was fun even in a humorous way, somewhat novel in concept, and you had to watch your six if you wanted to win. What did they do with the sequel many years afterwards ? Shall we settle on forgettable ?...

 

A good bunch of years ago (read many ) I was browsing through the shelves of a gaming store when I spotted a game that boasted several impressive awards for 'Best artificial intelligence' - the game in question was called Battle Isle II.

 

As I found out it had a lot to offer - if you didn't shy away from it just after reading the heavy-on-details manual that is (ranging from explaining logistics [fuel/refueling, ammo/reloading], weather [i.e. crossing an area of frozen water may prove fatal when the weather suddenly heats up] earning ranks from combat experience, unit repairs at specific spots or by special repair units, ground [wheeled and tracked vehicles], sea and air warfare, roads, hills, railroad tracks, unit stats and peculiarities, and general strategic notions...)

 

I spent many a good turn being pummeled by a well organised enemy, cursing my carelessness but enjoying the splashy faux 3D combat sequences while plotting my revenge - next turn ! (usually a few minutes later... after the dozens of enemy units carried out their instructions).

 

::

 

It was fun, and even awesome for the time. Fast forward to 2006 and what have you got ? Mostly a cartelized group of publishers afraid to put out anything on the shelves that may even come close to being qualified as cerebral. "Deep" is a dangerous word as well, so scratch that. Shoot and run, now there's our cup of tea; mainstream, baby! Instant gratification, shake your 'money-maker'!

 

Live with it ? Right now, I'm afraid so...

[Matri]

Agreed.

 

EDIT: I only used Wolf3D as an example. I loved it, and I just LOVED the bonus level in Doom2

[Thorondor]

Just spotted this GameSpy article disclosing the workings of F.E.A.R.'s AI at GDC '06. A small behind-the-scenes look...

 

::

 

Enjoy.

 

 

p.s.: chalk one more up for illusion++

[Slaughter]

Just spotted this GameSpy article disclosing the workings of F.E.A.R.'s AI at GDC '06. A small behind-the-scenes look...

Very interesting that! I have F.E.A.R., and did play it a little. It was killer for my laptop however (almost at the level of Doom 3), so I haven't played it much. What I saw from the AI was fairly impressive, even if it was just a skilful illusion

[Matri]

Well, in the end, isn't that what the whole thing is about?

[Slaughter]

Well, in the end, isn't that what the whole thing is about?

Until Skynet comes out of hiding, I guess it is

[Praetoris]

All I have to say is:

Creating a Virtuall Mind is not an easy thing ; )

 

Something I've been longing for lately though, except for the loss of retarded AI mentioned above, is an EVOLVING opponent. I've seen a couple examples of this but nothing on any grand scale yet. In Homeworld 2 for instance, the AI will start building ships effective against those I send towards him (if he survives that is ). It's a small thing, but always something

[Matri]

Until Skynet comes out of hiding, I guess it is

Here's a scary piece of news: There's a courier service called SkyNet over here in this part of the world. I did a double-take when I saw the sign.

[Thorondor]

On a little parenthesis of like mind to this thread's mood: Mr. Warren Spector speaks of what's in store for us if we don't mind our games, and then some, in his "Gaming At The Margins, Part 4" article on The Escapist online mag.

 

In a way he echoes the views of many gamers faced with an ungrateful question: What - is this the end of the road already ?

 

::

 

"What would we want games to become in the last six years of their existence?

 

If the world were snuffed out tomorrow, would we be satisfied that our last games had cooler explosions or that we created a more compelling simulation of criminal life?"

He sure hasn't all the answers, but the big questions seem right enough for me.

[Gimli]

Haven't noticed this posted before, so here goes:

 

While skimming through a PC magazine today, I noticed a little section of an article devoted to this. Could it be that finally something might change? Does anyone know anything about this? It doesn't surprise me that I haven't heard of this before, since all you here in the media is about the new shiny graphics and physics. Not that I mind those, but they get too much attention, compared to things that are actually useful for gameplay.

 

I apologize for the late post, but as you can see, it was unavoidable, and I think it was worth it.

[Crazy Gringo]

I heard of it as an idea when I surfed around the web for info on the Ageia PhysX processor....Not as a specific product....

One more piece of hardware to stuff into a otherwise cramped computer I guess....

Why don't they just made a customizable PCI-card that the user can program to handle one or more of the following tasks: Sound, graphics, AI, Physics and what ever else you might think of.

They are talking about making the graphics processors exchangeable like CPU's today so why not that customizable PCI-card?

[Gimli]

Knowing a bit about the complexity of building these in general, I would say that for the moment this won't be anyone's focus, as it would require quite a bit of research, and I would assume they have other priorities right now.

 

As for changing different types of processors, that would seem even more difficult, if not too complicated for now. As I understand it, each type of processors (graphics, sound, general etc.) works in a different way, to a degree, so it may not be all that possible right now, physically. Of course, I may be wrong, you should ask someone who knows more about electronics. Although I do have electronics subjects I have to take in my faculty, they're not my interest, so I only force myself to learn enough to pass the exams.

[Bomb Bloke]

The problem with a dedicated chip is that you're bound by what it can do, whereas when you're working in software mode you can do what you like.

 

Video cards are a near identical example. Sure they make things go faster, but if you want to include some special effect, you need to be sure the video card knows how to do it. If it doesn't, then you either don't use that effect, or you go back to software mode.

 

Most people have video cards these days, but I doubt many people have even heard of the AI chip. So even if you did write a game that supported it you'd need to include your own software AI code for those people who didn't have it. That means more work for developers, so I can see the technology holders paying software designers to use this chip until it becomes popular...

[Zeno]

Interesting thread, this. Advanced AI is a complex field in computing, and not really conceivable in computer games. Think about it-- Chess is a simple game in that there are a limited number of possible moves permitted in any turn. Yet it took, what, 2 super computers and decades of work to beat the top human player? When you have a computer game with hundreds of possible moves, any sort of advanced computer behavior becomes impossible due to the exponential growth in cause/effect calcuations.

 

Humans have innately reasonable "guesses" and react intelligently based on experience. Computers don't have these advantages; every decision is based on calculations. Advanced AI uses case modeling and pattern-learning logic, but it is still a far cry from human capability.

 

The best AI we might be able to hope for is something based on expert systems. Expert systems rely on the accumulated knowledge of humans in a particular subject, put into a data warehouse with optimized search/retrieval algorithms. This is still outside the scope of computer games; however, an automated database with cause/effect pattern analysis and a number of scripted responses based on human players could conceivably appear to be "better" AI.

 

I would love to see some advances in this field, but I think business deadlines and market competition make it unreasonable to expect from computer games. This is one reason why multiplayer-capability is demanded by most game players. Once a human figures out the game AI, the game becomes unchallenging. Multiplayer capability allows a game to have much longer playability.

 

------

 

Something to consider: Instead of advanced AI, games are often shipped with scripted behaviors. Rather than demanding developers to spend months making good scripts, games could be shipped (as some are) with good-quality script editors. This allows humans to make the scripts better, allows communities to grow around script-writing, and improves the game over time as the scripts are improved.

 

As game developers learn the advantages of open source code, or providing professional modding tools, I think we'll find improvements in not just AI, but many aspects of gaming.

 

--Zeno

[Alienated]

The A.I. programming is incredibly hard. Nonetheless, there are certain problems that have no existing algorythmic solutions by principle.

 

Generally I am satisfied with the A.I. of games. My problem usually is that the game rules and the A.I. are often dissonant and this is the error of both the A.I. programmers and the game designers. I believe that the main job of a game designer is to find out game rules (game mechanics) that require less A.I. programming to provide a still challenging and entertaining game.

 

For example the combats in RPGs have almost zero A.I. programming. The player has a variety of equipment, skills, and power-ups against a variety of monsters. The game designers just find out a plot, attach a combat system and voila, we get a cool game. The RPG is a classic and evergreen genre, RPG fans almost never curse the A.I. because there is actually none. The best A.I. is the non-existent A.I. This is the golden rule No. 1.

 

Tactical games have situations where A.I. programming is needed, and here comes the dissonance. The A.I. programmer can't salvage a bad game design if it is better to walk up to the target and release an autoshot in touch than aiming and shooting from ten metres. It is a common error in turn based tactical games that walks take relatively less action points than shots. The average game designer's aspect is this: the player should not move the figures a tile per turn because the gaming time might increase by "tiresome" field movements. The problem could be solved by good game design but the average game designers lack the talent to do that because the average game designers don't like to play video games, they just do it for money. (They should play their crappy games for the rest of their lives as due punishment.) In response, the A.I. programmer creates enemies who do very weird things on the battlefield to beat the player at all cost. I often can't decide whether I should laugh or curse when I am fated to watch my A.I-stricken enemies act.

 

And this is another common error in A.I. programming: the enemy always wants to win. By this aspect, a potential winner enemy is equivalent to the challenge. The enemy never gives up, never makes peace, never flees. And the A.I. programmer does everything so the enemy could not just try to win but win. As I said A.I. programming is hard, the A.I. compensates this by impertinent cheating including the knowledge and calculation of the players' forces and location as if there was no secret in war. And this is a great let-down in elements of tactical features. There is no need to bar enemy scouting and to risk leaving weak points here and there. The A.I. always knows where and what to attack, as soon as you leave a weak point, the enemy will attack it. This is an easy test to see how much the A.I. cheats. In real war, the commander made weak points look like strong points and strong points look like weak points, or just made risks, every commander had weak points here and there and prayed the enemy not to discover them.

 

I believe the best working A.I. lets the player win if the player uses the proper tactics, I mean the A.I. programmer has to make the A.I. intentionally weak against certain tactics. But it seems the opposite happens. The A.I. programmers make the computer enemy invulnerable against banal tactics to increase challenge. And the result will be that the player wins by tactics that belong in the world of madness. Actually the player does not beat the enemy in the game but the A.I. programmer, he wins by a way the A.I. programmer did not expect.

 

Complex A.I. thinking-calculation creates holes in the system. Let me take an example from Warcraft II. The enemy usually has a gold mine nearby its castle. When it is mined empty, the peasants search for new gold mine because more cash means more troops. Then you have to do one thing: build a guard tower nearby a gold mine or deploy some archers in the path to the gold mine and kill the peasants. The enemy will spend all its money on peasants, and when no more peasants are coming, you can decimate the enemy because it is flat broke. You can do that to oil tankers as well. If the A.I. were dumb enough to do nothing but execute the peasants and create combat units as many as the cash allows and wait for the final battle, then the player might have a harder time to win. So actually the cleverer A.I. makes the game easier (after the player has found the hole in the system, which wasn't too hard for me for one). Actually towers are the most dangerous units in Warcraft II in spite of that they are just rooted in place and fire at occasional targets.

 

And multiplayer mode was recommended as a solution for bad A.I. programming. Come on! People play games because they are lonely. And an unexperienced human player is often worse than the A.I. enemy.

 

All in all, the A.I. issues should be solved but not by mere A.I. programming but better game design too.

 

For example the designers could encourage the players to actually help the enemy do trivial things that make the enemy more an opponent. Some things that seem trivial to a human can also be very difficult to program properly. For example the player should be able to tell the enemy not to do this and that stupid thing. Just an example: what if you could tell the alien "do pick up your rifle you dropped in panic and fight like a soldier" in X-COM games? It is very difficult to program the aliens to search for their dropped weapons or even scavenge freely on the battlefield. Safe scavenging is a piece of cake for a human player whereas the mere idea makes me tired in the aspect of the computer-alien side. Aliens unable to scavenge seem very dumb.

 

Still, you have to admit AI's evolved from the days of Wolfenstein 3D and Doom, where they just point them at the player and shuffle forward, oblivious to walls and obstacles.

This A.I. tactics can be quite effective regardless of its simplicity (or rather because of its simplicity). If you are out of ammo or low HP, the bravely rushing enemy defeats you. If the enemy would camp here and there, then you could always retreat safely for power-ups. The "Can I play daddy?" level is the easiest mode. If you raise difficulty, then you suffer increased damage and meet more opponents, resulting often low HP and poor ammunition and the dumb rushing enemy becomes frightening. If they rush when you are out of ammo and camp when you have ammo and aim at the door, then the A.I. surely cheats by knowing your power and position.

 

One thing I've seen in lots of FPS games through the years is their idea of "good AI". On easy level the accuracy of the enemy is very low. They hit nothing, and you can run circles around them. Then, as you increase difficulty, they get much better. In quite a few cases they go inhuman, and make headshots from distances that no human player could ever manage, NO MATTER HOW GOOD HE WAS.

Even Duke Nukem 3d simulated the imperfection of bullet trajectory. This means you have to be extremely lucky to make perfect headshots from fair distances. No human can do that of course. The almighty God can do that. Lucky Luke can do that. But only a cheating A.I. can do that in reality.

[Thorondor]

*dusts off cobwebs*

 

Whatever Happened to Video Game AI? It's the title of the article at IGN.

 

::

 

All hail the rise of in-game Terminators... not.

[Thorondor]

*pops in again some two years later...*

 

https://www.rockpapershotgun.com/images/14/oct/alienisodlc1.jpg

 

RPS' Michael Cook writes about The Lost Future Of AI - part 1.

[NikiAlex]

My humble experience in RTS and TBS tells me this:

There are a quite a big number of conditions for AI to look for in order to be “good” (like comparison of battle stats, defense bonuses, current strength, detection and engagement ranges, weapon type, allies & enemies nearby, etc)

It is more of a challenge from design perspective to properly list all important conditions, behaviors’ and well balanced reaction thresholds. Once this is set up as schemas and functions then programming tackles begin to kick-in.

There more conditions and situation you push to implement the higher the change some condition will go wrong, will be missed at all and this of course will lead to annoying in-game situation. For sake of the discussion let’s assume conditions and responses were set up top notch and you have AI character/unit that appears “intelligent„ and deliver to you nice amount of challenge. Congratulations you have excellent AI for you Character class/Unit. But wait a minute… I have so many different character/unit classes, special traits and so many weapon types. Yeah, I will need to customize AI for most of them.

Now we have some awesome AI let’s do some combat in RTS. Ops AI need to calculate most of his condition every frame in order this to be really efficient. We have a CPU performance issues. OK then let’s reduce condition checking to every second, but not every frame. Well it appears this is not frequent enough. AI is not as good as we wanted.

At some point you are asking, how much time do I need to invest to got my good AI opponents

Just my two cents.

Will be happy to read more hands-on on this topic

Cheerz

[NoXTheRoXStaR]

AI is never as good as we want it to be. It's either too moronic, or too abrasive. There's never that fine line in which we say..."yeah that's the kind of AI I prefer". I honestly believe it's trial and error, any AI can be dealt with once you figure out a way to manipulate it's faults and then take advantage of its said faults. If you created the most complex AI (built for use of gaming) and despite all of the brain maneuvering, calculated risks and manipulations you can't progress? It's no longer it's a game, it's better than you are, and you had best not introduce it to your wife, because it'll fuck her better than you ever would. So...there ya go.