Forum

Life-Like AI vs Good AI

Discuss Artificial Intelligence and Bot programming.

Moderator: InsideQC Admins

IS Life-Like AI Better Always? Post a Reply with reasoning, please.

Yes
3
27%
No
8
73%
 
Total votes : 11

Postby Spirit » Tue Sep 14, 2010 6:25 am

Didn't we have this discussion about Monsters before? Maybe it was at func.

I think the "bad AI" in Quake is actually part of its interesting gameplay. Because you can learn patterns. Think of Super Mario with "good AI", I don't think it would be much fun. Quake is an arcade-type game, if you like it or not. The monsters might be boring (at least by now) but I think the right way for new monsters would be to think of creative new patterns, not random behaviour (or crosshair evasion like in Nehahra, god, what a bad decision).
Improve Quaddicted, send me a pull request: https://github.com/SpiritQuaddicted/Quaddicted-reviews
Spirit
 
Posts: 1025
Joined: Sat Nov 20, 2004 9:00 pm

Postby frag.machine » Tue Sep 14, 2010 11:55 am

I don't think the dumb monster AI was intentional. I can't conceive as a proposital decision a monster unable to follow a player from one room to another, or that ignores their comrades corpses while patrolling an area. Romero talked a lot about how the monsters in the original Quake concept would be fewer than in Doom, because they would be more challenging. There are reasons for this, and none was something like "hey, let's make all monsters dumb and predictable".

First, it was a case of balancing the scarce available CPU resources between monster AI and the render engine. Better AI requires things like A* pathfind or the stack-based target tracking we can see in the Frikbot - in other words, more memory and CPU cycles, when the target was a Pentium 100 with 8MB of RAM.

Second, the entire single player and coop aspects of the game were overlooked because Quake always was all about mutiplayer over TCP/IP.

And finally, let's not forget that Quake was a project developed under an internal crysis at id; people were really pissed off with Romero acting like a jerk and ignoring his duties in the project, and this spilled over the gameplay in many aspects.

In any case, the final result wasn't in any way what they intended to do.
I know FrikaC made a cgi-bin version of the quakec interpreter once and wrote part of his website in QuakeC :) (LordHavoc)
User avatar
frag.machine
 
Posts: 2053
Joined: Sat Nov 25, 2006 1:49 pm

Postby Spike » Tue Sep 14, 2010 1:05 pm

Quake never used TCP/IP.
The actual networking supports UDP/IP, SPX/IPX, serial, serial-with-modem.

Quake was kinda rushed towards the end, yeah, the sort of time when things like AI gets polished and stuff. The AI is fairly simple, and just works as-is. Heck, monsters still try to attack you even during intermission!

The monsters are fun and interesting regardless of their AI level. Sure, its a weak point, but it doesn't stop it from being a fun arcade game, it merely breaks the immersion a little.
They all have different quirks, like grunts having a worse aim if you're moving, ogres with different weapons for different distances with grenades bouncing off walls behind you, fiends with their bouncy jumpyness of death, zombies that require you to reserve some rockets, flying things with slow projectiles that encourage you to strafe lots, shamblers and the shambler dance, the annoying vore missiles (grr), and, urm... well, knights, but enough said about those. Basically, Quake is a dancing game - think of the AI as the music, and the monster type as the tempo.
Artificial Stupidity can work just as well as Artificial Intelligence, for many things.
Spike
 
Posts: 2874
Joined: Fri Nov 05, 2004 3:12 am
Location: UK

Postby frag.machine » Tue Sep 14, 2010 2:01 pm

Spike wrote:Quake never used TCP/IP.
The actual networking supports UDP/IP, SPX/IPX, serial, serial-with-modem.


Sorry, I used "TCP/IP" to refer the protocol stack (although Quake protocol was indeed developed using TCP/IP over DSL/LAN and later ported to UDP/IP according Carmack himself).

Spike wrote:Quake was kinda rushed towards the end, yeah, the sort of time when things like AI gets polished and stuff. The AI is fairly simple, and just works as-is. Heck, monsters still try to attack you even during intermission!


There's nothing wrong in liking Quake as is, as long people don't believe it was something intended to be this way.

Spike wrote:The monsters are fun and interesting regardless of their AI level. Sure, its a weak point, but it doesn't stop it from being a fun arcade game, it merely breaks the immersion a little.


And, for a game that was intended to be all about immersion and sensorial experience, it's a train wreck. Luckly, it has enough good strong points to overcompensate this and provide a really long and pleasing experience for us.
I know FrikaC made a cgi-bin version of the quakec interpreter once and wrote part of his website in QuakeC :) (LordHavoc)
User avatar
frag.machine
 
Posts: 2053
Joined: Sat Nov 25, 2006 1:49 pm

Postby mankrip » Wed Sep 15, 2010 5:16 am

Quake was supposed to be like Doom 3, only with flying dragons and Daikatana's "RPG elements".



I may or may not have been kidding in this post.
Ph'nglui mglw'nafh mankrip Hell's end wgah'nagl fhtagn.
==-=-=-=-=-=-=-=-=-=-==
Dev blog / Twitter / YouTube
User avatar
mankrip
 
Posts: 908
Joined: Fri Jul 04, 2008 3:02 am

Postby frag.machine » Wed Sep 15, 2010 12:07 pm

mk wrote:Quake was supposed to be like Doom 3, only with flying dragons and Daikatana's "RPG elements".



I may or may not have been kidding in this post.


Huh ?! Quake == closet monsters everywhere and no duct tape to wrap the flashlight to the shotgun ? :shock:
I know FrikaC made a cgi-bin version of the quakec interpreter once and wrote part of his website in QuakeC :) (LordHavoc)
User avatar
frag.machine
 
Posts: 2053
Joined: Sat Nov 25, 2006 1:49 pm

Postby Team Xlink » Tue Jan 04, 2011 1:46 am

Sajt wrote:
#2 - Maybe you can fool idiots (common riff-raff game players, average IQ in double digits) by upping hitpoints, but I don't make games with them in mind. (It was immediately apparent that Halo was being made precisely for this audience, though.) Also, enemies with lots of hitpoints are a lame console/arcade-ism. We should be able to do better than that.
.


I agree this, is a rather cheap way of creating "intellegent" AI, it usually comes accross as, either "Well, the players a god-like person, so we have to have huge numbers of enemies to kill, but because of the large number they can't be too smart because then the player isn't "god-like", or "Well our AI is way to easy, lets make it harder, hey lets give them more health so they last longer".

Some games it works good, because the your supposed to be an all powerful person, however in others it comes across as cheap and pointless.

Sajt wrote:#4 - This is only if you want every single part of the game to come through on the first playthrough. I HATE AI behaviours made obvious. One enemy type running away everytime another is killed does not give the illusion of intelligence, unless you thought Diablo's Scavengers were smart. Also, in Fear the guys yell "I see his flashlight!" They are just showing off, and it is especially lame the second time and every time after. If a game should be replayable there should be behaviours (and emergent situations) that don't come out so often. Also, obvious behaviours make it easier for the player to piece together the monster logic, like we've all done ages ago in Quake, which lets us beat each monster reliably using the same pattern. It ruins the whole idea of intelligence or unpredictability, and you lose the fear or respect of the monsters, and the whole game universe.
.


I agree, they need to not be a "played them once played them a thousand times" type of scenario, singleplayer relies either on, length or replayability, usually... With length unless your fighting the monster sparingly or only in one section, it usually comes across as, "darn it these guys again, they take more time then tis worth" it looses the sense of involvement, and makes it a chore instead of a game. (Contrarying achievements do the same, but that hasn't stopped a lot of people (including myself) from playing a game to its fullest to get that last achievement thats a pain and a chore. Humans, in general never let you know how they are fully, if they did, it get boring... Game AI should be the same.

Sajt wrote:#5 - It's a legitimate point. More complex AI will become totally arbitrary to the player, and he will either assume it's totally random, or he will acknowledge that he'll never learn the patterns. But, I think that this is the right direction in the future. Human enemies shouldn't be predictable. Maybe on the third time through a level, something will happen that will make the player think to himself, "you know, it almost seemed like that guy got fed up with his buddies and ditched them". If that happens a few times, you're in the right place, I think.


Yes, but sometimes certain human enemies are always predictable, they always on Aerowalk, run to the rocket launcher grab the green armor, rocket jump to the ai, grab the lgihting drop through the whole to the ramp below jump to the mega health, stalk in a pattern then repeat. This makes it so you can generally wait for them to come around the "track" again and kill them. But some human enemies are always doing something different, it comes across as, "There is no routine in this guys gameplay" which makes it harder to get a good strategy against him.

Sajt wrote:#6 - Yeesh, I still haven't been able to appreciate the designer who is not a programmer or the programmer who is not a designer. But this must be a fault of mine.


I can't say to much on this point, I don't have enough experience to feel a point in sharing my opinion.

Sajt wrote:#11 - You better be slick at implementing this! It reeks of Electronic Arts sports or racing games with their rubber-band "catch up" AI cheating behaviour.


AI cheating behavoir such as this is annoying, its like, the guys a lap down he can't catch up, yet he is right there a minute later.

Sajt wrote:#13 - I totally sympathize with this one! It's unfortunate.


Can't really say on this one, I haven't read the article fully.

Sajt wrote:#14 - NO! Okay, maybe if you are making a dead-linear game, but I'm not interested in that... I have never been entertained by setpieces and scripts. THESE break suspension of disbelief faster than apparent bugs do. They tell you that you're in a movie, not a game. DOWN WITH "CINEMATIC" EXPERIENCES!!


Scripts make a one time play through sort of game, in games that have enemies surprise you, it might work the first time but then it becomes, "yep hes around the corner in a wall when I cross this sidewalk a noise plays to distract me and it comes out." Half-Life did this with the headcrabs a good amount.

Sajt wrote:#16,17,18 - This is basically waypointing, so it's okay, as long as the playtesters are creative and rigorous.


Waypointing can be good except when they follow it exactly, even when shooting at you they stay on there path.

Sajt wrote:#19 - I saw screw friendly NPCs altogether! Or let the player kill them if he wants.


Yes, if I shoot a rocket or snipe a npc in the head, they should die, regardless of whether there helping you or not, it ruins the experience.

Sajt wrote:#20 - Ehh. I guess so, but I rarely enjoy this when playing the game.


Same as #13, I can't really say on this one as I haven't fully read the article.

Sajt wrote:#21 - This should be rule number one (not "focus on the ILLUSION of intelligence", which possibly contradicts it)


Yes, the illusion of intellegeince is a bad goal, it needs to be intellegent not an illusion of it.

Sajt wrote:#26 - A repeat of #4. I see their point, but somehow I can't fully accept it. Maybe because the little of Halo I did play was not fun at all. But I know that in games like Thief, the illusion of smart AI (which there wasn't, but a new player assumes so) made the game supremely stressful and not-fun... Maybe the key is moderation. Or good difficulty settings. Or else ease the player in: easy (fun?) at the beginning and hard (fun?) at the end. Like Super Mario Bros 3. Decades later I still haven't beaten it, but it's still fun to replay.


I'm in the same opinion in general as you are on this one.

Sajt wrote:#32 - This needs to be subtle. Otherwise it's about as enjoyable as getting gold-star stickers and being told you're special. (As in, counterproductive.)


Thats annoying isn't it? "Give your self a pat on the back", you say that everytime we do what were *supposed* to do.

mh wrote:Agreed with most of the above, although I haven't analysed it or the original points in much detail yet. All the same though, I can't help thinking that a game that achieves what it sets out to do is successful on it's own terms, and that the market is a lowest common denominator anyway.


Yes the market seems to be the lowest common denominator, its making a good game that matters, but tell that to the publisher...

mh wrote:What annoyed me about something like Half Life AI (to change things a little) was that it was so damn repeated. Remeber when the black ops ninja-like guys attacked? The first time you were pinned down behind a crate and they were circling around you, trying to flush you out with grenades, and so on? The first time that happened it was cool. The second time I thought "oh, this again". The third time I said "fuck it" and god-moded through. The fourth time it didn't even bother me because I had left god mode on.


The same thing happened to me.

mh wrote:That's not enjoyable, where's the player reward in that? Contrast with the set-piece in Doom 3 where the floor dops, you get the Berserker, and about 30 dumb zombies come at you. It's silly, it's contrived, there's no AI to speak of, but mowing through a room full of the lowest enemy class while invincible gave immense player satisfaction.


Yes, because it was fun and your a god like figure at that point, wheres most of the game, your not.

mh wrote:So I suppose the point is that AI designers should always always always keep this one question in the back of their minds: "how is this going to help the player enjoy the game more?" If AI isn't working towards that goal - within the parameters of the game - then it's failed, no matter how realistic or life-like it might be.


That is also true.

mk wrote:The monster infighting actually happens a lot less in Quake than in Doom, and is a feature that not only makes the available gameplay strategies more diverse (getting the monsters to kill each other may take more time than killing them directly, but it allows us to economize our own ammo), but also gives them more personality, since it gets pretty obvious that they're so angry and insane that they won't forgive anything, not even from their own mates.


Yes, but it gets a little old on the second time throughj=.

mk wrote:The AI in Half-Life can really get pretty annoying. It works great for when the scenery gives us some tactical options that requires some thinking to beat them, but it's very, very annoying when our only option is to crawl as slow as possible to not let more than a few pixels of their head shows up for us to shoot them with the sniper. By the last stage I got so bored of crawling all the time that I cheated through the rest of the game. Soldier of Fortune and Turok 2 also suffers from this "crawling-to-snipe syndrome".


Yes, you need to have not only options for different play styles, but options that are enjoyable for those play styles.

mk wrote:I may suck at aiming, but having that as the only tactical option for people who can't aim like a bot is ridiculous. I'm a lot better moving than shooting, maybe that's why I prefer Sauerbraten over any other FPS online; it's the only one I can always have a good time with, even if I'm on the bottom of the scoreboard.


Being able to have a good time despite being the lowest on the rankings, is rather enjoyable, because your enjoying the game even if your not doing good.

mk wrote:About Sajt's comment on #4, I think that the way to prevent obvious behaviors from getting predictable is to only make them obvious after their actions has been taken. This way the player can still learn why he lost when he's beaten, and in doing so the player may perceive and understand how the different reactions of the AI works, but he still won't have means to predict which of those possible reactions the AI will choose to do next time.


Yes, but then instead of the 2 time being the time when things get old, the third time is and so on. I think its better to always have a little bit of unpredictability no matter how many times you've played them, but not fully unpredictable.

mk wrote:IMO the main problem with (both AI and human) team communications in gaming is that it's usually implemented through a different interface (exclusive menus, console commands or keybindings), instead of being implemented through the actual mechanics of the gameplay. For example, why not have a team radar (either in the HUD or in a render-to-texture display in some areas) to always inform the position of your teammates, instead of having them disrupt their focus from the gameplay to either type messages or choose pre-defined ones to inform this? Communication should get the point across through the gameplay, instead of disrupting it.


Yes, stopping to type in game is rather annoying at times and some times by the time you finish typing, what you were going to say was pointless, because they already capped. A radar is a good idea, but that still doesn't describe the situation enough if there were two or three of your team mates, but the one enemy, had quad, mega health and red. Voice chat has solved this a good amount but it removes some of the immersion.

Electro wrote:Just to post again quickly, while all good and valid points (for the most part).. I believe Xlink was actually posting regarding bots. :)


My first post was regarding bots, but this AI discussion about monsters, has proven very useful for my AI endeavors.

Sajt wrote:Yeah, he was! Good to get that point out there. BTW, spectating bots or monsters is great for examining AI.


I've never really thought of that, I'm defiantly going to be doing a lot of spectating from now on.

Sajt wrote:Anyway, about infighting, that's another one of those things that seems to make the game more fun, while at the same time destroying immersion by making monsters retards with opaque logic. I always take advantage of infighting when I'm playing a custom map with early shalraths. But could this situation (leading a shalrath bomb around) be fun without the monsters being dumb? Previously the monster you lead the bomb into is totally unaware of it until he's hit, and when he's hit he instantly knows who threw it (even if you led around several corners) and predictably goes after the shalrath. If, instead, he tried to avoid the bomb while it was flying around, that could be fun. You could clear a path by riskily leading a bomb around. And if you are lucky you could still lead it into another monster. (But if you did, that monster would probably blame you, not the shalrath.)


Then would be fun, it would make them smart and insane at the same time.


Sajt wrote:Then again, I'm pretty sure all Quake's monsters were supposed to be blindly insane and completely stupid (it would seem out of place for them to be a bit more rational). But when you play the game, that doesn't come across, what comes across is bad AI. You could make them scream all the time and foam at the mouth, but I would get really annoyed by that. In any case, I don't think stupid monsters makes for a very oppressive atmosphere.


Yes, but it is an arcade-like experience as it is, the multiplayer shines the most. Even though the stupid AI makes it a rather different experience then what a smart AI would provide, it wouldn't be as "Quakeish"

Spike wrote:screw lifelike ai, make efficient ai, 60 bots, and give them teamplay capabilities, get them to hunt you in packs.


That would be difficult to play but it would be fun to see the team work used.

Spike wrote:your bot/monster/monkey should not act too stupidly. it shouldn't keep trying to bash the square peg in the round hole. having said that, if its a bot then it would be pretty lifelike if it did that...
bots that don't target players that are sitting idle is a matter of courtesy, but you can't always distinguish that from a player sniping.
A lifelike AI will favour killing whoever it has a grudge against.


Yes, that is true, it even leads to some mindless team killing but that is rather fun at times. But having my partner in a singleplayer game randomly turn and shoot me in the head killing me would provide a pretty "WTF" moment, but it'd be fun/funny none the less.

Spike wrote:For QuakeWorld, a hard AI is one that camps the items properly. But if it always stands on the same spot, you'll know where to lob that grenade.


Yes, QuakeWorld AI is good that not only does smart things but does smart things the smart way.



Spike wrote:If you want to keep a player entertained, you have to keep the AI changing and using different tactics and just generally dynamic.


Yes, dynamic gameplay is entertaining, but without at least a small amount of "pattern-ish-ness", it becomes random, but you did say "generally dynamic".

Spike wrote:But remember, this is a game, and thus the goal should never be truely lifelike AI, the goal should be fun AI.
The player should feel that he has a chance at beating it, but it shouldn't be too easy.


If a game it isn't fun its failed.

frag.machine wrote:
Spike wrote:The player should feel that he has a chance at beating it, but it shouldn't be too easy.


That is true, it makes it not a mindless killing charge, but not a hopeless battle either.

Spike wrote:Nor too smart, like the DJQuake's dandy shambler that politely asks the player to examine his shotgun. :lol:

I haven't played DJQuake's so I don't really know about that.

Spike wrote:But returning to the original question: no, none of the examples given by Team XLink fits in AI behavior, either good or bad, in my opinion - those are just small behavior tweaks to not piss off human players. The bot didn't reasoned, in any way, it was unfair to attack or pursuit someone because he/she is less skilled or have a bad internet connection or is afk. Remember, by definition a bot is something without emotions or moral values. It don't give a damn if you were in disadvantage.


Yes, but thats were the Bot verses Life-Like AI comes in, theres a difference.

Spike wrote:Good AI IMO would be a bot realize that I always follow the same route into the map, navigating from one item spawn to another, and use this information to ambush me in some way; or conclude, after discharging the thunderbolt underwater, that it was not a good movement and should be avoided in the future; or yet, to learn to avoid or circumvent a tossed grenade when sees it.


Yes that would be intelligent AI, and would be fun and skilled, but rather difficult at times.

Spirit wrote:Didn't we have this discussion about Monsters before? Maybe it was at func.

I think the "bad AI" in Quake is actually part of its interesting gameplay. Because you can learn patterns. Think of Super Mario with "good AI", I don't think it would be much fun. Quake is an arcade-type game, if you like it or not. The monsters might be boring (at least by now) but I think the right way for new monsters would be to think of creative new patterns, not random behaviour (or crosshair evasion like in Nehahra, god, what a bad decision).


Yes, Quake wouldn't be as Quake-ish if it didn't have the "bad ai" It is Quake.

frag.machine wrote:I don't think the dumb monster AI was intentional. I can't conceive as a proposital decision a monster unable to follow a player from one room to another, or that ignores their comrades corpses while patrolling an area. Romero talked a lot about how the monsters in the original Quake concept would be fewer than in Doom, because they would be more challenging. There are reasons for this, and none was something like "hey, let's make all monsters dumb and predictable".

First, it was a case of balancing the scarce available CPU resources between monster AI and the render engine. Better AI requires things like A* pathfind or the stack-based target tracking we can see in the Frikbot - in other words, more memory and CPU cycles, when the target was a Pentium 100 with 8MB of RAM.

Second, the entire single player and coop aspects of the game were overlooked because Quake always was all about mutiplayer over TCP/IP.

And finally, let's not forget that Quake was a project developed under an internal crysis at id; people were really pissed off with Romero acting like a jerk and ignoring his duties in the project, and this spilled over the gameplay in many aspects.

In any case, the final result wasn't in any way what they intended to do.


I don't really know the history of id software all too well, but the target machine did have a limiting factor on terms of what they could do. And like you said Quake's about multiplayer.

Spike wrote:Quake never used TCP/IP.
The actual networking supports UDP/IP, SPX/IPX, serial, serial-with-modem.

Quake was kinda rushed towards the end, yeah, the sort of time when things like AI gets polished and stuff. The AI is fairly simple, and just works as-is. Heck, monsters still try to attack you even during intermission!

The monsters are fun and interesting regardless of their AI level. Sure, its a weak point, but it doesn't stop it from being a fun arcade game, it merely breaks the immersion a little.
They all have different quirks, like grunts having a worse aim if you're moving, ogres with different weapons for different distances with grenades bouncing off walls behind you, fiends with their bouncy jumpyness of death, zombies that require you to reserve some rockets, flying things with slow projectiles that encourage you to strafe lots, shamblers and the shambler dance, the annoying vore missiles (grr), and, urm... well, knights, but enough said about those. Basically, Quake is a dancing game - think of the AI as the music, and the monster type as the tempo.
Artificial Stupidity can work just as well as Artificial Intelligence, for many things.


Yes, it works well in Quake, Quakes, Quake as we know it. Stupid AI did work in a sense in this case, we all still play Quake right? Maybe not for the same reasons, but we still play.
Team Xlink
 
Posts: 368
Joined: Thu Jun 25, 2009 4:45 am
Location: Michigan

Postby Sajt » Tue Jan 04, 2011 4:35 pm

Sorry if I'm responding to something I've already responded to (I forget).

Also keep in mind that I don't care one bit about online multiplayer... even though that was the topic of this thread. Bot AI is a totally different school than Monster AI. The former has to simulate a human at a keyboard, and has to either be as good as a real human or cheat to make up the difference. I'm not interested in that at all. Monster AI is about simulating an actor on its own terms.

And in terms of singleplayer AI I'm actually less interested in making a game "fun" by way of arcade-like mechanics, as I am in establishing a mood and then not breaking it. Quake had arcade-like mechanics, which were good except they wouldn't work in a modern high-poly game without stylizing the graphics (which would destroy the mood). That conundrum kind of sucks.

But when it comes to monster AI, Quake failed in my opinion. Or at least, it's not fun anymore when you've killed 200,000 ogres in your time. Or even 200 ogres. When playing, the monsters in Quake are reduced in my head to a small handful of logic routines. They'll never surprise you, and you know for certain that they'll never do anything new. They cease to be intimidating beings and they become puppets, about as interesting as the enemies in a monster-avoid pattern game like Super Metroid or Mario.

An oppressive or mysterious atmosphere depends on the possibility of something unexpected happening. As soon as you "see the matrix", the whole thing becomes dull (and philosophically depressing!). So how do you avoid this? You could write more AI routines that included a lot of stuff like "if (random() < 0.01)". But if you're actually interested in advancing the art of game AI, you'd try something else. Now, I'm totally going off on a limb here, because I haven't tried any of these things, but... if AI wasn't comprised of simple if/else routines, but of something more compartmental like neural networks or even genetic routines, you could possibly randomly molest one or more of these components to add infinite varieties of slightly corrupted logic which would result in new gameplay behaviours. I'm not sure if this sort of thing would work for human actors, or just robot monsters, but it still would be an interesting path to try.

Also, note that in all my posts in this thread I'm just thinking aloud. I will often say that "atmosphere is more important than 'fun' mechanics". But this theory is unverified. It's true that the ultimate game I envision is very serious and very realistically-treated, and tries its hardest to not seem like a game at all. But if this game were ever made, I can't guarantee that I would play it more than I would play something like Super Mario.

Also, there are two directions that games are headed in. Well, three.

One is the old-school "true" game. These are comprised of mechanics that are intended to be balanced. Atmosphere comes second. In these games, the developers don't think twice of teleporting your companion NPC to you if he gets stuck. RTSes and stats-driven RPGs are this way. They're like boardgames with fancy graphics. Basically every game from the previous century fall under this classification, except some failed high-concept attempts like Ultima Underworld, and inane dull-a-thons like SimCity.

Second is the "sandbox" game. This is apparently where I'm headed. They forget about balance completely, and just let the player loose. Far Cry is like this, though it's a terrible game in my opinion. Morrowind is like this. Grand Theft Auto series as well, to a smaller degree. I think there are a lot more games like this these days but I wouldn't know. But none of these games come close to its potential. In these games you can afford to put the atmosphere, or the mood, or the "world" first. Fun is usually derived from the player's freedom. Probably more challenging is keeping him from getting bored if he wants to keep heading in one direction.

The third type is hardly worth mentioning, though it makes up the vast majority of games. These are the assembly-line linear games made by people who don't understand video games as a new creative medium. They have scripts. They often put the "world" first, but they're written not as games but as movies in which the viewer is pestered to keep pressing buttons.

mh wrote:Agreed with most of the above, although I haven't analysed it or the original points in much detail yet. All the same though, I can't help thinking that a game that achieves what it sets out to do is successful on it's own terms, and that the market is a lowest common denominator anyway.


Speaking for myself, I'm not that interested in making a game that is successful on its own terms. I guess I'm more of an engineer at heart than a game developer. I want to try new or better ways of doing things. I also don't care about the market at this point. That said, I'm also not interested in making crappy "programmer demos" of new AI concepts. Oh well, we'll see if I get anywhere in my lifetime.

mh wrote:That's not enjoyable, where's the player reward in that? Contrast with the set-piece in Doom 3 where the floor dops, you get the Berserker, and about 30 dumb zombies come at you. It's silly, it's contrived, there's no AI to speak of, but mowing through a room full of the lowest enemy class while invincible gave immense player satisfaction.


That is fun, yes, but the player recognizes it as a wink in his direction by the game developers. A wink which signifies, "we don't take this very seriously", as they throw a goofy/campy scenario at you. Depending on your vision, you may not want to compromise the "seriousness" of your game atmosphere in this way.

Your example doesn't represent all kinds of "fun", just one, and I don't think it's essential. I would say it's more amusing than essentially fun.

Team Xlink wrote:
mk wrote:IMO the main problem with (both AI and human) team communications in gaming is that it's usually implemented through a different interface (exclusive menus, console commands or keybindings), instead of being implemented through the actual mechanics of the gameplay. For example, why not have a team radar (either in the HUD or in a render-to-texture display in some areas) to always inform the position of your teammates, instead of having them disrupt their focus from the gameplay to either type messages or choose pre-defined ones to inform this? Communication should get the point across through the gameplay, instead of disrupting it.


Yes, stopping to type in game is rather annoying at times and some times by the time you finish typing, what you were going to say was pointless, because they already capped. A radar is a good idea, but that still doesn't describe the situation enough if there were two or three of your team mates, but the one enemy, had quad, mega health and red. Voice chat has solved this a good amount but it removes some of the immersion.


In general you should avoid different input "modes", yes. Fortunately for me I don't have much interest in how to solve the problem of communication with bots in a FPS.

Voice chat to me is interesting (though I've never used it). If the game is sci-fi or military and you play as a guy with a radio (built into your helmet maybe), the voicechat could represent that. It could even be disrupted by nearby electro-magneto-stuff or by world objects like big concrete walls. In a low-tech game (maybe medieval) it would be pretty amusing to have voicechat attenuated like any other sound. The point is voicechat shouldn't exist "on top" of a game, but should be built in to the game mechanics.

Team Xlink wrote:
Sajt wrote:Yeah, he was! Good to get that point out there. BTW, spectating bots or monsters is great for examining AI.


I've never really thought of that, I'm defiantly going to be doing a lot of spectating from now on.


I used to spectate the bots in Tactical Ops (UT mod). I'm sure more games support this as well.

Team Xlink wrote:
Sajt wrote:Then again, I'm pretty sure all Quake's monsters were supposed to be blindly insane and completely stupid (it would seem out of place for them to be a bit more rational). But when you play the game, that doesn't come across, what comes across is bad AI. You could make them scream all the time and foam at the mouth, but I would get really annoyed by that. In any case, I don't think stupid monsters makes for a very oppressive atmosphere.


Yes, but it is an arcade-like experience as it is, the multiplayer shines the most. Even though the stupid AI makes it a rather different experience then what a smart AI would provide, it wouldn't be as "Quakeish"


I don't care one bit about Quake's multiplayer. The singleplayer to me is interesting because it suggests a game that is so much better and more immersive and atmospheric than it actually is. It's like they did the first 10%, and left it up to you to imagine the rest. If I actually had an imagination, I could enjoy Quake a lot more by imagining the rest while I played it. But I have to suffice with trying to come up with a game design that is "quake done right".

Of course, you can't significantly improve Quake's AI without redesigning the levels, and possibly more. The levels are made and balanced around the existing AI. The ideas I get from Quake point towards a completely new game.

Spike wrote:screw lifelike ai, make efficient ai, 60 bots, and give them teamplay capabilities, get them to hunt you in packs.


This is an example of what I was just talking about... If this were done in Quake singleplayer, the entire level would empty itself and the monsters would attack the player simultaneously. Unless some monsters were also supposed to be guarding some artifact, or doing something else. Or if there were multiple factions of monsters that didn't like each other.

And of course in coop this would become very interesting. Quake coop is awful. (Also in part because of the level design.) Improved AI, and probably more wide-open spaces, would make coop so much more fun.

Team Xlink wrote:
Spike wrote:But remember, this is a game, and thus the goal should never be truely lifelike AI, the goal should be fun AI.
The player should feel that he has a chance at beating it, but it shouldn't be too easy.


If a game it isn't fun its failed.


This is true if you have to beat a certain group of enemies to beat the level/game. If you're in an indoor level, and these monsters are blocking a hallway, it's one thing. If you could find another way around them, or activate a trap or distract them, it's another thing. Of course that also counts as "beating" them.

I'm thinking about "sandbox" games again of course. By the way, I mentioned earlier that "mysteriousness" (the factor which keeps the player immersed in the world rather than in a game) depends on there being the possibility that new unbeforeseen things will happen. Another way to keep it would be to have optional monsters/bosses that are WAY more powerful than the player. Unless they are the size of suns, they would still be beatable, but extremely difficult (like the Ruby/Emerald weapons in Final Fantasy 7). Bonus points if you have a sadistic save system, because the player would probably not even dare to try to kill them. (Just kidding... actually no bonus points for a sadistic save system.)

Spirit wrote:Didn't we have this discussion about Monsters before? Maybe it was at func.

I think the "bad AI" in Quake is actually part of its interesting gameplay. Because you can learn patterns. Think of Super Mario with "good AI", I don't think it would be much fun. Quake is an arcade-type game, if you like it or not. The monsters might be boring (at least by now) but I think the right way for new monsters would be to think of creative new patterns, not random behaviour (or crosshair evasion like in Nehahra, god, what a bad decision).


I answered this in my other ramblings in this post. It's true that Quake is arcadey, and that my ideas and preferences go away from Quake. I'm not as interested in the arcadeyness of Quake as in the potential of its mood/atmosphere.

By the way, I don't remember Nehahra that well but didn't Quoth have crosshair-dodging? I seem to remember it in Warpspasm, especially with the rocket guys on the roof. (Or am I thinking of a different map?) It was really annoying though kind of amusing once you realized what they were doing.

Anyway, as for coming up with new patterns, that's an idea for someone who wants to do it, and it could yield good results, though I'm not personally interested. To me it's moving sideways instead of forward.

I don't condone adding "randomness" (that is, completely arbitrary behaviour) to AI. Instead what I'm thinking of is more like adding randomness or complexity to their decision making. Perhaps they even operate with more than one goal or task at any given time, and if decision making (which controls which goals/tasks are at which priority) is more complex, you get more and more novel behaviours, but they're still reasoning.

Team Xlink wrote:Yes, it works well in Quake, Quakes, Quake as we know it. Stupid AI did work in a sense in this case, we all still play Quake right? Maybe not for the same reasons, but we still play.


The thing is, I don't play Quake much anymore. (Singleplayer that is. I never played the multiplayer much at all.) After the gazillionth time, it's just not there. Not that there was much there when I first played it, but I was more naive then!

In conclusion, I'll reiterate that almost all my ramblings are irrelevant to a discussion about bots and deathmatch...
F. A. Špork, an enlightened nobleman and a great patron of art, had a stately Baroque spa complex built on the banks of the River Labe.
Sajt
 
Posts: 1215
Joined: Sat Oct 16, 2004 3:39 am

Postby daemonicky » Thu May 05, 2011 11:25 pm

Game is a model of reality. And I guess this simplification is why I enjoy them and not jumping on rooftops, looting houses, burning villages or raping women (there are some japanese simulations/games) ... Each such simple simulation can offer me some interesting insights or allow me to try something forbidden without getting myself killed (jumping from rooftops, riding Formule 1), blamed or convicted (buring, killing, raping).

OP was talking about perfect AI. Well, it depends how you define perfect. If "perfect" like it will always win or perfect like fitting generic player needs. You can probably simulate this "being nice to people with lag" with computer.

To my knowledge, first dungeon games were copies of paper games and paper games had simplistic XP system. Then Elder Scrolls make this system realistic.

Quake AI is Quite simplistic and it is enough. Pac Man AI is very simplistic But FEAR had nice opponents. I guess it is like with art, "each person likes something else".
User avatar
daemonicky
 
Posts: 185
Joined: Wed Apr 13, 2011 1:34 pm

Postby Sajt » Fri May 06, 2011 6:29 am

daemonicky wrote:Game is a model of reality.


Wrong! Though most nowadays game developers would agree with you rather than me...
F. A. Špork, an enlightened nobleman and a great patron of art, had a stately Baroque spa complex built on the banks of the River Labe.
Sajt
 
Posts: 1215
Joined: Sat Oct 16, 2004 3:39 am

Postby daemonicky » Fri May 06, 2011 6:41 am

Sajt wrote:
daemonicky wrote:Game is a model of reality.


Wrong! Though most nowadays game developers would agree with you rather than me...


Why wrong?

- I might have oversimplified. There are some abstract games which are not models of reality, like tetris. But, who knows, there might be some process in reality which tetris models ...

To me, most of the games are simulations, usually of single hero (fps, rpg, adventure) or group of people (sims, strategy games).

- Also fantasy games might not apply here. We do not have magic any more. :)

- I am also wrong because if I take game as element of game theory, then any game is not necessarily model of our reality.

Or because of some other reason?
User avatar
daemonicky
 
Posts: 185
Joined: Wed Apr 13, 2011 1:34 pm

Postby leileilol » Fri May 06, 2011 7:13 am

WE PLAY GAMES TO ESCAPE REALITY AND HAVE FUN not be stricted in reality restricting the fun

this is a point todays western game developers have forgotten about.

realism is overrated

play Operation Flashpoint and tell me how much fun you had with it
i should not be here
leileilol
 
Posts: 2783
Joined: Fri Oct 15, 2004 3:23 am

Postby behind_you » Sat May 07, 2011 8:17 am

leileilol wrote:WE PLAY GAMES TO ESCAPE REALITY AND HAVE FUN not be stricted in reality restricting the fun


agreed. :D
User avatar
behind_you
 
Posts: 237
Joined: Sat Feb 05, 2011 6:57 am
Location: Tripoli, Libya

Previous

Return to Artificial Intelligence

Who is online

Users browsing this forum: No registered users and 1 guest