depends on what you consider "smart" - creating challenging ai isnt difficult - every simple chess program does it...simply programm it to respond to every action with the optimal counter action - et voila challenging battles...but that has nothing to do with ai...
if youre talking about "real" ai, you talk about the ability to create and adapt - especially outside of battle - and that to me is mostly marketing... what they probably will do is have simply scripted events like
every hour mobs gain 1 mob, once they reach 10 mobs, attack the nearest village...things like that...
You don't understand anything about programming. A chess program is easy. You can easily see how they work by calculating every possible move several moves deep, and evaluating the end result. A chess board is static, with a very refined set of rules, which makes it easy. The chess computer isn't responding to you in any meaningful way, it is just analyzing the new position.
actually, being a programmer, i dare to disagree.
it is far easier to make a kickass AI based on rules than you may think. i have allready posted this earlier on, but its apperant that people dont take the time to read the thread, but just dive in with invalid opinions disguised as truth.
To be honest, I loved the Trinity; specific roles for everyone so they could master it. I think that AI could still use the Trinity; however, if the healer heals; it will take some time or even kiting / distraction to get the boss back onto the tank.
Tera has a good idea of this until the Lancers (tanks) get the aggro buff (glyph) and then it's too easy.
depends on what you consider "smart" - creating challenging ai isnt difficult - every simple chess program does it...simply programm it to respond to every action with the optimal counter action - et voila challenging battles...but that has nothing to do with ai...
if youre talking about "real" ai, you talk about the ability to create and adapt - especially outside of battle - and that to me is mostly marketing... what they probably will do is have simply scripted events like
every hour mobs gain 1 mob, once they reach 10 mobs, attack the nearest village...things like that...
You don't understand anything about programming. A chess program is easy. You can easily see how they work by calculating every possible move several moves deep, and evaluating the end result. A chess board is static, with a very refined set of rules, which makes it easy. The chess computer isn't responding to you in any meaningful way, it is just analyzing the new position.
actually, being a programmer, i dare to disagree.
it is far easier to make a kickass AI based on rules than you may think. i have allready posted this earlier on, but its apperant that people dont take the time to read the thread, but just dive in with invalid opinions disguised as truth.
Everybody and his brother says he is a programmer on the internet!
I am not a programmer, but I am a chess player. So, I know how these programs work.
They are nothing even remotely close to what I would call advanced AI. It is just an example of computing power.
They have had chess computer programs that could play at a 2000+ ELO, Master level, for like 30 years. Current day programs are estimated at about 3000 ELO, thus putting them at above human level.
The OP doesn't have the faintest idea of what he is even talking about. Academic AI and game AI are not the same thing. Further, more advanced AI systems than "holy trinity" (which is not even AI by the way, it's a simplistic system that can't even considered to have real behavior) have existed for almost a decade.
F.E.A.R. (which came out in 2005, a mere YEAR after WoW) for instance uses action planning, which means the agent only knows the end goal, not any of the steps to accomplish it. That is EONS more advanced than "trinity AI" and wasn't built by MIT. And we've had almost 10 years of AI advancements since then.
MIT has jack to do with game AI advancements, and they can hardly solve a problem they have never bothered to attempt solving, making this guy's entire point both wrong and moot.
This post is laughable and the OP's severe ignorance of anything to actually do with the topic of game AI is readily apparent.
He has all the credibility of a crack junkie asking to hold on to your wallet.
It's a question of difficulty and enjoyment. A 'smart' AI would simply focus your healer or maybe it's so advanced they could look at your HP pool, armor, analyze your skill rotation and decide you are the weakest or slowest person and focus you.
For a casual PVE player, this would segregate them from the content or they would simply not play the game; resulting in a low population. The game would be more frustrating than fun.
It is a lie because NPC's cannot think,they ONLY react to the scripts you give them.It is simply PR tactics that i don't agree with,keep it real and keep it the truth or go home.
Another question that made me cringe when they were asked about what type of gamer they were catering too and of course,you know the answer before Georgeson even opened his mouth.We cater to EVERY gamer.
I have seen some cases where PR guys will give it to you straight,but it is rare,they would steal the money right out of your pocket if they could.
Never forget 3 mile Island and never trust a government official or company spokesman.
It is a lie because NPC's cannot think,they ONLY react to the scripts you give them.It is simply PR tactics that i don't agree with,keep it real and keep it the truth or go home.
They can 'think' in a way if you give them likes, dislikes, preferences, choices, etc. That's what the story bricks system does. They will react because everything in the world is tagged and can be analyzed by the NPCs.
Read up on how the storybricks system works and get a better understanding of what they are trying to achieve.
It is a lie because NPC's cannot think,they ONLY react to the scripts you give them.It is simply PR tactics that i don't agree with,keep it real and keep it the truth or go home.
They can 'think' in a way if you give them likes, dislikes, preferences, choices, etc. That's what the story bricks system does. They will react because everything in the world is tagged and can be analyzed by the NPCs.
Read up on how the storybricks system works and get a better understanding of what they are trying to achieve.
Storybricks is script, it just has more "yes and no" checks then dumb AI's.. As for NPC's remembering things, that isn't new.. It's called faction, and in many games in the past you could have neutral faction and get no help from NPC's in a fight, but if you were allied for example and in danger, a NPC would come to your aid if you were in range.. That is not rocket science.. If you keep track of every groups faction, which original EQ had many, it's not that hard to have a deep AI similar to Storybricks..
ALL mobs in any game have 2 choices.. They are given random orders, or they follow a script of yes/no code checks.. This applies to combat as well.. Either the mob will randomly attack anyone in a given range, or it will follow code instructions.. So devs can spew all the BS they want about smart AI actions, but once that code is discovered, all their effort is meaningless then.. Players will find out what the code is and abuse it and use it..... So removing taunt does nothing in the end.. UNLESS there are more random mobs then they are leading on.. And if that is the case, then it's not smart AI, it's Random AI and we have Zergfest..
The OP doesn't have the faintest idea of what he is even talking about. Academic AI and game AI are not the same thing. Further, more advanced AI systems than "holy trinity" (which is not even AI by the way, it's a simplistic system that can't even considered to have real behavior) have existed for almost a decade.
F.E.A.R. (which came out in 2005, a mere YEAR after WoW) for instance uses action planning, which means the agent only knows the end goal, not any of the steps to accomplish it. That is EONS more advanced than "trinity AI" and wasn't built by MIT. And we've had almost 10 years of AI advancements since then.
MIT has jack to do with game AI advancements, and they can hardly solve a problem they have never bothered to attempt solving, making this guy's entire point both wrong and moot.
This post is laughable and the OP's severe ignorance of anything to actually do with the topic of game AI is readily apparent.
He has all the credibility of a crack junkie asking to hold on to your wallet.
You dolts realize that the whole MIT thing was just sarcasm right?
Originally posted by Rydeson Originally posted by arcatomOriginally posted by WizardryIt is a lie because NPC's cannot think,they ONLY react to the scripts you give them.It is simply PR tactics that i don't agree with,keep it real and keep it the truth or go home.
They can 'think' in a way if you give them likes, dislikes, preferences, choices, etc. That's what the story bricks system does. They will react because everything in the world is tagged and can be analyzed by the NPCs.Read up on how the storybricks system works and get a better understanding of what they are trying to achieve.http://massively.joystiq.com/2011/08/08/storybricks-opening-the-pandoras-box-of-mmo-design/It's not just a set script. Storybricks is script, it just has more "yes and no" checks then dumb AI's.. As for NPC's remembering things, that isn't new.. It's called faction, and in many games in the past you could have neutral faction and get no help from NPC's in a fight, but if you were allied for example and in danger, a NPC would come to your aid if you were in range.. That is not rocket science.. If you keep track of every groups faction, which original EQ had many, it's not that hard to have a deep AI similar to Storybricks..
ALL mobs in any game have 2 choices.. They are given random orders, or they follow a script of yes/no code checks.. This applies to combat as well.. Either the mob will randomly attack anyone in a given range, or it will follow code instructions.. So devs can spew all the BS they want about smart AI actions, but once that code is discovered, all their effort is meaningless then.. Players will find out what the code is and abuse it and use it..... So removing taunt does nothing in the end.. UNLESS there are more random mobs then they are leading on.. And if that is the case, then it's not smart AI, it's Random AI and we have Zergfest..
I was following StoryBricks back when they were trying to make a game out of it. It's not random choices, it's a decision tree kind of thing. It's not all that different from how people decide to do things.
Given a set of "wants", the AI will follow the decision tree to achieve the goal. The example they use is a group of gnolls or something wanting a relatively quiet area, where a few travelers come through. If an area gets to busy, the gnolls will search out another area. If there aren't enough travelers to attack, they'll search out another area. By itself, that one set of gnolls isn't very intelligent.
However, the NPC guards use the same system. So players can report a gnoll sighting to the guards, and the guards may search out and drive the gnolls off, or they may offer a reward to players for killing the gnolls. Now the system is a little more intelligent and responsive to the players.
None of the NPCs will do something that the developers don't first supply an action for. The gnolls will not start writing poetry, but they may try to bribe players into leaving them alone, or they may hire assassins to kill a particularly aggressive player. The AI is more intelligent than current MMORPG AI systems, and it's certainly more responsive than existing AI systems.
The combat stuff is pretty trivial in comparison. We've had advanced combat AI for years now. We could have combat AI that wipes the floor with players by always making the right choices, and that includes group combat. Blizzard could do that in WoW. They need to be able to do stuff like that for testing purposes. They just don't turn that AI on the players because it wouldn't be fun and would be a road block towards the goal of dropping loot for the players.
I can not remember winning or losing a single debate on the internet.
I was following StoryBricks back when they were trying to make a game out of it. It's not random choices, it's a decision tree kind of thing. It's not all that different from how people decide to do things.
Given a set of "wants", the AI will follow the decision tree to achieve the goal. The example they use is a group of gnolls or something wanting a relatively quiet area, where a few travelers come through. If an area gets to busy, the gnolls will search out another area. If there aren't enough travelers to attack, they'll search out another area. By itself, that one set of gnolls isn't very intelligent.
However, the NPC guards use the same system. So players can report a gnoll sighting to the guards, and the guards may search out and drive the gnolls off, or they may offer a reward to players for killing the gnolls. Now the system is a little more intelligent and responsive to the players.
None of the NPCs will do something that the developers don't first supply an action for. The gnolls will not start writing poetry, but they may try to bribe players into leaving them alone, or they may hire assassins to kill a particularly aggressive player. The AI is more intelligent than current MMORPG AI systems, and it's certainly more responsive than existing AI systems.
You are expecting too much from the AI that spawns mobs
Originally posted by lizardbones
The combat stuff is pretty trivial in comparison. We've had advanced combat AI for years now. We could have combat AI that wipes the floor with players by always making the right choices, and that includes group combat. Blizzard could do that in WoW. They need to be able to do stuff like that for testing purposes. They just don't turn that AI on the players because it wouldn't be fun and would be a road block towards the goal of dropping loot for the players.
I agree. This has already existed, which is why it is PR hype. It isn't hard to make a computer better than players, but it would be stupid to turn it on, because no one wants to play a near impossible game.
Originally posted by jimprouner Originally posted by lizardbones I was following StoryBricks back when they were trying to make a game out of it. It's not random choices, it's a decision tree kind of thing. It's not all that different from how people decide to do things. Given a set of "wants", the AI will follow the decision tree to achieve the goal. The example they use is a group of gnolls or something wanting a relatively quiet area, where a few travelers come through. If an area gets to busy, the gnolls will search out another area. If there aren't enough travelers to attack, they'll search out another area. By itself, that one set of gnolls isn't very intelligent. However, the NPC guards use the same system. So players can report a gnoll sighting to the guards, and the guards may search out and drive the gnolls off, or they may offer a reward to players for killing the gnolls. Now the system is a little more intelligent and responsive to the players. None of the NPCs will do something that the developers don't first supply an action for. The gnolls will not start writing poetry, but they may try to bribe players into leaving them alone, or they may hire assassins to kill a particularly aggressive player. The AI is more intelligent than current MMORPG AI systems, and it's certainly more responsive than existing AI systems. You are expecting too much from the AI that spawns mobs Originally posted by lizardbonesThe combat stuff is pretty trivial in comparison. We've had advanced combat AI for years now. We could have combat AI that wipes the floor with players by always making the right choices, and that includes group combat. Blizzard could do that in WoW. They need to be able to do stuff like that for testing purposes. They just don't turn that AI on the players because it wouldn't be fun and would be a road block towards the goal of dropping loot for the players.
I agree. This has already existed, which is why it is PR hype. It isn't hard to make a computer better than players, but it would be stupid to turn it on, because no one wants to play a near impossible game.
Those aren't my expectations, those are the descriptions of the AI behavior from the StoryBricks developers and from SOE. They were working on the StoryBricks AI years before SOE ever started talking to them. It was originally a system for building procedural story content in a story based game world. That's just how the AI works. It doesn't just "spawn mobs".
I can not remember winning or losing a single debate on the internet.
“If I had asked people what they wanted, they would have said faster horses.” - Henry Ford
The argument that the AI doesn't exist has been debunked thoroughly. It's hilarious that people with no credentials in the field of game AI (or game design at all), made sweeping statements about what is possible or likely. It's pretty obvious that the people arguing against the possibility or likely-hood of advanced AI, have no interest to ever look beyond the trinity system. These people consider the trinity integral for MMORPGs. It has nothing to do with what is possible with AI.
So I'm going to go ahead and remind everyone why so many people want to do away with the trinity to begin with. The trinity forces every class to fit into one of three jobs: Tank, Healer, or DPS. Each class must do their job just as well as the next. Otherwise, in the difficult "endgame" content, some classes are clearly superior to others.
In WoW, a "hybrid" just means you can pick one of many different jobs to do. It does not mean you do multiple jobs with less efficiency simultaneously. They even decided to take out the 5% less dmg/healing/tank penalty that they put in to preserve the spirit of the hybrid. The reality is even 5% makes those classes less desirable.
Classic EQ on the other hand, didn't balance every class to do these jobs equally well. Clerics were better healers than druids and shamans (much better, in fact). Warriors were needed for tanking raid mobs. Hybrids did multiple things less well.
This caused classes like Clerics, Wizards, and Rogues to be highly sought after for raiding. Where as most other classes you wanted just 1-2 for their unique buffs. There was a limit to the use for multiple warriors, shamans, and enchanters for example. Paladins and SKs pretty much lacked a purpose in raiding at all. Ranger were strictly worse at high end content until there were morphed into a more pure DPS class in Luclin (not classic).
There is not much a place for certain class concepts in a trinity system. Such as "tanky dps." If you need an off-tank, you want another pure tank class; not some tanky hybrid. So "tanky dps" can be inherently overpowered (monks in velious), because they do just as much dmg as squisher dps classes, but have more survivability. Alternately, they may be underpowered (rangers until Luclin) because their tankyness isn't used and they do less DPS than dedicated DPS classes.
The trinity basically puts a tight cap on class versatility and customization. Everyone ends up being pure Tanks, Healers, and DPS for the endgame. Only when the trinity is lifted (in the form of PvP, for trinity games) do you see more in between.
Some people love "just tanking" or "just healing." Most people don't. Most people enjoy a bit more variety than this. Which is why EQN came up with the no trinity, horizontal progression model. To keep things interesting and customizable.
On somewhat of a side note: the trinity is bad for PvP. It encourages extremely powerful healing. If there are powerful heals, they should be on somewhat long cooldowns. Which is never the case in trinity games. There is no real way to set up ideal PvP gameplay when powerful heals with no cooldown have to be taken into account.
You can make it so the healers die fast, but that's lame for the healer.
You can make it so that you need to run healers oom, which is extremely boring.
You can make it so you need to chain CCs in order to actually kill people...this is the WoW system and most people hate it.
I would say healing is the main reason Arena in WoW is not popular as a spectator esport, or liked by the majority of players.
Yeah, I'm gonna have to pretty much disagree with most of your post here. First of all the trinity is not just tank/dps/healer. There is also cc. Like has been pointed out the original trinity was tank/dps/cc. The term trinity is now used generically for a system that has defined roles. It also doesn't "force" people into 3 roles. There is healing, dps, tanking, offtanking, CC, buffing and hybrids which can be a combination of roles but still unique in their own right in what they offer to a group. For example EQ had bards which didn't neatly fit into a category but were damn useful.
Only warriors could tank bosses? SK's were useless? Sorry but that is just wrong. We had a SK main tank in EQ that tanked the bulk of our raids and was one of the better tanks I have seen. Sure, some fights we used a warrior or a paladin but that was the beauty of the system. There were many possible strategies and the class system allowed for creative solutions and the ability to be successful with what you had rather than using the same strat/raid makeup that every other guild used. We also had some really great paladin tanks. We experimented with different things. Sometimes it worked and other times it didn't. That was the fun and was especially gratifying when we came up with our own method of doing something.
Classes in EQ weren't balanced to all be exactly equal because they had actual variety. Yes, clerics were the best healers but you know what? Our guild used druids on CH rotations in a pinch. Sure the 75% CH was not as good but stick a druid in between 2 clerics in the rotation and it worked. Not ideal but it worked. Druids weren't thought of as inferior clerics because they brought DPS and buffs and snares and roots that the cleric didn't have. This is true flexibility and variety. Not making everyone's abilities the same but with different animations/weapons/gear to give the illusion that you are doing something different.
I was following StoryBricks back when they were trying to make a game out of it. It's not random choices, it's a decision tree kind of thing. It's not all that different from how people decide to do things.
Given a set of "wants", the AI will follow the decision tree to achieve the goal. The example they use is a group of gnolls or something wanting a relatively quiet area, where a few travelers come through. If an area gets to busy, the gnolls will search out another area. If there aren't enough travelers to attack, they'll search out another area. By itself, that one set of gnolls isn't very intelligent.
However, the NPC guards use the same system. So players can report a gnoll sighting to the guards, and the guards may search out and drive the gnolls off, or they may offer a reward to players for killing the gnolls. Now the system is a little more intelligent and responsive to the players.
None of the NPCs will do something that the developers don't first supply an action for. The gnolls will not start writing poetry, but they may try to bribe players into leaving them alone, or they may hire assassins to kill a particularly aggressive player. The AI is more intelligent than current MMORPG AI systems, and it's certainly more responsive than existing AI systems.
The combat stuff is pretty trivial in comparison. We've had advanced combat AI for years now. We could have combat AI that wipes the floor with players by always making the right choices, and that includes group combat. Blizzard could do that in WoW. They need to be able to do stuff like that for testing purposes. They just don't turn that AI on the players because it wouldn't be fun and would be a road block towards the goal of dropping loot for the players.
WoW this thread is still alive and ongoing.. as much as i know i wrote something similar to lizard at page 5.. and with that the thread should basicly end.
And i want to add to your last paragraph ( and i guess i have even wrote that somewhere in this thread, but can't be bothered to read all 30 pages yet again).
The target of storybrick in combat is not to be a perfect AI, to make always the right dicisions, it is much more to make reasonable and sometimes even wrong(but reasonable) decisions, like a lot of humans do it every day.
That a group of mobs don't act like one men attacking the perfect target, but much more behave like a group, with every single one of them behave a little bit different, a littly bit asynchron, and do even other decisions. With other words, they try to give the mobs the ability to make mistakes.. and not either the perfect counter, or just one stupid move all of them do.
And storybrick is by the way not the first AI tool, which tried it.. anyone interested can look up about emotional AI and you can find some interesting read. (although it was never used in a MMO, at least as much as i know)
Upgrading AI with something like storybricks is a good decision all around, in theory.
The problem many people have is upgrading AI in such a way that it removes traditional role based combat, and the false notions that its 1) the only way to improve AI, 2) that its somehow the more logical alternative and/or 3) the logical evolution of game AI. These notions are an insult to the intelligence of players who have any understanding of game mechanics and artificial intelligence on even the most basic level. SOE's sudden hostility and outright mockery of the previous forms of video game AI (especially mmorpg AI), specifically threat management (ie. "the trinity"), has been nothing less than ridiculous and a shameless attempt to promote a new form of combat gameplay, not just advanced AI as they'd have you believe. The scary thing is, most of their staff are apparently drinking their own kool-aide, which became obvious during SOE live when no less than a dozen times I heard them refer to threat management and hate generating mechanics as something along the lines of "insulting a mobs mother." That is not a good sign, and I'm slightly embarrassed on their behalf.
It comes down to a matter of gameplay preference, not accepting the logical evolution of AI. Many MMORPG players, especially those who have enjoyed EQ, would prefer to see them build upon the foundation of threat management combat which promotes the established roles found in traditional rpgs. Some of which are more action based (tanks and some melee) and others which are more layed back or passive. This traditionally has allowed for all types of people to participate and enjoy mmorpgs, both those who enjoy being the leader with better situational awareness, aim and reflexes, as well as those looking for a role with less responsibility, less twitch capabilities, and frankly a role that isn't stressful (some people work for a living). SOE quite obviously intends to leave these combat mechanics behind altogether in favor of arcade action-adventure combat where all the "roles" (if you can call them that) are action based and, at one time or another, become the de factor "tank" by adapting to combat scenarios and utilizing movement abilities like jumps, dodges, rolls or teleports. This combat has proven fun in some games, but has always been zergy in massively MULTIPLAYER online rpgs, hence the resistance from the more experienced, less naive mmo player.
I, for one, am not fool by this, nor am I subverted by their juvenile attempts to mock the AI that has served games well for decades (and will continue to serve), regardless of how trivial it has become in modern MMOs. The triviality of modern MMOs should be attributed not to the code, but the philosophies industry leaders have employed to generate a larger playerbase and make their game more all-inclusive ($$$). Fortunately for SOE, the vast majority of their target audience hasn't the slightest concept of how mmos work, or should work, so they stand to attract quite a playerbase with their grand claims and hyperbolic remarks.
Upgrading AI with something like storybricks is a good decision all around, in theory.
The problem many people have is upgrading AI in such a way that it removes traditional role based combat, and the false notions that its 1) the only way to improve AI, 2) that its somehow the more logical alternative and/or 3) the logical evolution of game AI. These notions are an insult to the intelligence of players who have any understanding of game mechanics and artificial intelligence on even the most basic level. SOE's sudden hostility and outright mockery of the previous forms of video game AI (especially mmorpg AI), specifically threat management (ie. "the trinity"), has been nothing less than ridiculous and a shameless attempt to promote a new form of combat gameplay, not just advanced AI as they'd have you believe. The scary thing is, most of their staff are apparently drinking their own kool-aide, which became obvious during SOE live when no less than a dozen times I heard them refer to threat management and hate generating mechanics as something along the lines of "insulting a mobs mother." That is not a good sign, and I'm slightly embarrassed on their behalf.
It comes down to a matter of gameplay preference, not accepting the logical evolution of AI. Many MMORPG players, especially those who have enjoyed EQ, would prefer to see them build upon the foundation of threat management combat which promotes the established roles found in traditional rpgs. Some of which are more action based (tanks and some melee) and others which are more layed back or passive. This traditionally has allowed for all types of people to participate and enjoy mmorpgs, both those who enjoy being the leader with better situational awareness, aim and reflexes, as well as those looking for a role with less responsibility, less twitch capabilities, and frankly a role that isn't stressful (some people work for a living). SOE quite obviously intends to leave these combat mechanics behind altogether in favor of arcade action-adventure combat where all the "roles" (if you can call them that) are action based and, at one time or another, become the de factor "tank" by adapting to combat scenarios and utilizing movement abilities like jumps, dodges, rolls or teleports. This combat has proven fun in some games, but has always been zergy in massively MULTIPLAYER online rpgs, hence the resistance from the more experienced, less naive mmo player.
I, for one, am not fool by this, nor am I subverted by their juvenile attempts to mock the AI that has served games well for decades (and will continue to serve), regardless of how trivial it has become in modern MMOs. The triviality of modern MMOs should be attributed not to the code, but the philosophies industry leaders have employed to generate a larger playerbase and make their game more all-inclusive ($$$). Fortunately for SOE, the vast majority of their target audience hasn't the slightest concept of how mmos work, or should work, so they stand to attract quite a playerbase with their grand claims and hyperbolic remarks.
I, however, will not be among them.
Wow... a triple straw man...straw men rather. Nice to create your own false notions and then shoot them down...
1. "The only way to improve AI"... who, other than you is saying that there is only one way forward?
2. "it's somehow the more logical alternative" ... it's an alternative...it's different... more logical? Not sure that's the point. More like a player would behave in that situation? Yeah, that's a bit closer.
3. "The logical evolution of game AI" ... there you go with that "logical" thing again.
What in fact is happening here, and in other places--EQN is not by any means the only MMORPG in development attempting new mob combat behavior--is a recognition that things have been done in one specific way for a long time and that doing them in a different way might be more fun.
So they're trying a new way. And this new way involves attempting to have mobs and friendly NPCs behave in a manner that is more realistic... more like how they would behave if there was a player controlling them instead of just a canned script.
I won't bore you with the technical details--I've already linked those in previous posts in this thread. Suffice it to say, that this system also has tweakable difficulty and the people designing it are fully aware that the bottom line is enjoyment and satisfaction.
This new way will be different than the traditional trinity-based combat with almighty tank aggro kludges. Whether it will be more enjoyable remains to be seen.
But I for one am looking forward to seeing this new way of PVE combat. God knows that when MMOs became big business, innovation and risk taking took a back seat to the tried and true financially successful formulas. It's nice to see a team with the cojones to try to do something different. That alone makes me inclined to cut them a lot of slack and just wait and see what they come-up with.
"Social media gives legions of idiots the right to speak when they once only spoke at a bar after a glass of wine, without harming the community ... but now they have the same right to speak as a Nobel Prize winner. It's the invasion of the idiots”
― Umberto Eco
“Microtransactions? In a single player role-playing game? Are you nuts?” ― CD PROJEKT RED
Streetfighter 2 has the hardest AI. Play on level 8 and you won't be able to beat the second fighter, but the first one is always possible to win. The first enemy of level 8 in arcade mode in a fighting game is the type of difficulty i want for any game. You don't have to be a pro to win, but you still have to be careful.
The OP doesn't have the faintest idea of what he is even talking about. Academic AI and game AI are not the same thing. Further, more advanced AI systems than "holy trinity" (which is not even AI by the way, it's a simplistic system that can't even considered to have real behavior) have existed for almost a decade.
F.E.A.R. (which came out in 2005, a mere YEAR after WoW) for instance uses action planning, which means the agent only knows the end goal, not any of the steps to accomplish it. That is EONS more advanced than "trinity AI" and wasn't built by MIT. And we've had almost 10 years of AI advancements since then.
MIT has jack to do with game AI advancements, and they can hardly solve a problem they have never bothered to attempt solving, making this guy's entire point both wrong and moot.
This post is laughable and the OP's severe ignorance of anything to actually do with the topic of game AI is readily apparent.
He has all the credibility of a crack junkie asking to hold on to your wallet.
You dolts realize that the whole MIT thing was just sarcasm right?
"Coming from the software development world I know for a FACT that SOE is not employing people from MIT's AI department for their game. This would be the only way to produce AI even close to challenging enough to make the "Trinity" system obsolete. However, even MIT hasn't created AI advanced enough to make this claim." - Bearknight
Well, we either have someone who has a very unique take on the OP's message, or bearknight just outed his alt.
There isn't a "right" or "wrong" way to play, if you want to use a screwdriver to put nails into wood, have at it, simply don't complain when the guy next to you with the hammer is doing it much better and easier. - Allein "Graphics are often supplied by Engines that (some) MMORPG's are built in" - Spuffyre
Seems to be a lot of confusion between AI and difficulty. The two are not mutually exclusive. AI can be tweaked to be more difficult, but I don't think that is the primary intention of SoE. They are using it to make a living world. Mobs will react differently based on the scenario. A group of rogues might only show themselves to solo adventurers or small groups where they think they have the advantage. More importantly the NPCs can interact with each other, making you feel like you are in a world rather than a movie set.
This is deja vu all over again.. I remember all the hype how the 4th pillar in SWTOR was going to change the MMO genre forever.. blah blah blah.. We all saw how over hyped that shit was.. If you enjoyed single player games like KoTOR or Dragon Age, then SWTOR was great.. but for a mmo experience it failed.. Anyways.. I'm hearing the same hype about this AI bs too.. How it's going to be the next great thing to change the MMO world .. blah blah blah.. Fool me once, shame on you.. Foll me twice, shame on me.. I will NOT bite that bait and run with it..
Originally posted by Rydeson This is deja vu all over again.. I remember all the hype how the 4th pillar in SWTOR was going to change the MMO genre forever.. blah blah blah.. We all saw how over hyped that shit was.. If you enjoyed single player games like KoTOR or Dragon Age, then SWTOR was great.. but for a mmo experience it failed.. Anyways.. I'm hearing the same hype about this AI bs too.. How it's going to be the next great thing to change the MMO world .. blah blah blah.. Fool me once, shame on you.. Foll me twice, shame on me.. I will NOT bite that bait and run with it..
If anything, the 4th pillar was viewed here with a decent amount of skepticism. On the AI, it's an alternative to the most common form of combat and mob AI. If MMOs adopt it, that's a pretty significant change, primarily because most MMOs are centered around combat.
There isn't a "right" or "wrong" way to play, if you want to use a screwdriver to put nails into wood, have at it, simply don't complain when the guy next to you with the hammer is doing it much better and easier. - Allein "Graphics are often supplied by Engines that (some) MMORPG's are built in" - Spuffyre
SOE accumulated quite some knowledge about AI. Vanguard has advanced AI. They just switched it off a few weeks after release. Back then mobs ran away and came back with friends for example. EQ2 has some quite good mercanaries.
Both games had to tune down AI because of performance problems. The EQN engine is better performing, allowing the CPU to compute more AI.
SOE accumulated quite some knowledge about AI. Vanguard has advanced AI. They just switched it off a few weeks after release. Back then mobs ran away and came back with friends for example. EQ2 has some quite good mercanaries.
Both games had to tune down AI because of performance problems. The EQN engine is better performing, allowing the CPU to compute more AI.
They had to tune it down because people don't really want hard fights, they want it to be predictable.
Mobs ran in eq as well ... in 1999. It's not a matter hardware performance, most player now could not accept a real challenge.
Originally posted by zymurgeist The "advanced" AI exists. It has always existed. Chess programs are more complicated by far than any RPG AI. Hobbyists used to write routines for AI in basic in the late seventies. Sony won't be using it for the same reason all the other companies who claim to have had it haven't. It's extremely CPU intensive. Great for single player games but way too slow and expensive for a MMO.
Blizzard did an interview, or wrote an article several years ago talking about the combat AI. I'm paraphrasing, but they basically said writing a good combat AI would be easy. The AI could have all the abilities that a player would have, it would be situationally aware and wouldn't get excited and make mistakes. It would use the ideal combination of abilities for any given situation. They need that kind of AI for testing.
They didn't use the more advanced AI because it wouldn't be fun for players. The AI was too good to be used against players. It would be good for testing though.
SOE is doing something completely different with the Story Bricks AI. That gnoll will attack because it is angry with you for killing his brothers. Or the gnoll will move to a less populated part of the world so it is easier to attack adventurers. The combat AI is going to be a version of that kind of behavior. It won't make perfect decisions, but if it's attacking someone, and another player heals that someone, it'll attack the healer instead of making the right decision at the beginning of the fight and attacking the healer right away. Something like that anyway. Or even better, mobs in groups will act like they are a group, repositioning themselves, healing each other and possibly running away.
The group AI has been done years ago in single player games. At least a single player game. Groups of mobs would have a leader and they would have the ability to do things as a group. If the leader was killed, the group would be disorganized unless another one of the mobs moved into the position of leader, and they would again act as an organized group.
"Intelligent" AI isn't smoke and mirrors, and it's not science fiction. It is well within the realm of what can be done. The only important part is that is has to be fun. Keep in mind that mobs don't exist to replace players, they are there to provide a challenge, and ultimately drop loot.
+1 The AI is out there, it just can't be used as it wold 'win' every time.
@OP I understand your frustration, but the points you bring up are largely half truths and inaccurate. Sadly, your points on the trinity are accurate and I hope someone makes it so a balanced group is required to complete group content.
If you want a new idea, go read an old book.
In order to be insulted, I must first value your opinion.
Comments
actually, being a programmer, i dare to disagree.
it is far easier to make a kickass AI based on rules than you may think. i have allready posted this earlier on, but its apperant that people dont take the time to read the thread, but just dive in with invalid opinions disguised as truth.
over 20 years of mmorpg's and counting...
To be honest, I loved the Trinity; specific roles for everyone so they could master it. I think that AI could still use the Trinity; however, if the healer heals; it will take some time or even kiting / distraction to get the boss back onto the tank.
Tera has a good idea of this until the Lancers (tanks) get the aggro buff (glyph) and then it's too easy.
Everybody and his brother says he is a programmer on the internet!
I am not a programmer, but I am a chess player. So, I know how these programs work.
They are nothing even remotely close to what I would call advanced AI. It is just an example of computing power.
They have had chess computer programs that could play at a 2000+ ELO, Master level, for like 30 years. Current day programs are estimated at about 3000 ELO, thus putting them at above human level.
The OP doesn't have the faintest idea of what he is even talking about. Academic AI and game AI are not the same thing. Further, more advanced AI systems than "holy trinity" (which is not even AI by the way, it's a simplistic system that can't even considered to have real behavior) have existed for almost a decade.
F.E.A.R. (which came out in 2005, a mere YEAR after WoW) for instance uses action planning, which means the agent only knows the end goal, not any of the steps to accomplish it. That is EONS more advanced than "trinity AI" and wasn't built by MIT. And we've had almost 10 years of AI advancements since then.
MIT has jack to do with game AI advancements, and they can hardly solve a problem they have never bothered to attempt solving, making this guy's entire point both wrong and moot.
This post is laughable and the OP's severe ignorance of anything to actually do with the topic of game AI is readily apparent.
He has all the credibility of a crack junkie asking to hold on to your wallet.
It's a question of difficulty and enjoyment. A 'smart' AI would simply focus your healer or maybe it's so advanced they could look at your HP pool, armor, analyze your skill rotation and decide you are the weakest or slowest person and focus you.
For a casual PVE player, this would segregate them from the content or they would simply not play the game; resulting in a low population. The game would be more frustrating than fun.
Take the Magic: The Gathering 'What Color Are You?' Quiz.
It is a lie because NPC's cannot think,they ONLY react to the scripts you give them.It is simply PR tactics that i don't agree with,keep it real and keep it the truth or go home.
Another question that made me cringe when they were asked about what type of gamer they were catering too and of course,you know the answer before Georgeson even opened his mouth.We cater to EVERY gamer.
I have seen some cases where PR guys will give it to you straight,but it is rare,they would steal the money right out of your pocket if they could.
Never forget 3 mile Island and never trust a government official or company spokesman.
They can 'think' in a way if you give them likes, dislikes, preferences, choices, etc. That's what the story bricks system does. They will react because everything in the world is tagged and can be analyzed by the NPCs.
Read up on how the storybricks system works and get a better understanding of what they are trying to achieve.
http://massively.joystiq.com/2011/08/08/storybricks-opening-the-pandoras-box-of-mmo-design/
It's not just a set script.
Storybricks is script, it just has more "yes and no" checks then dumb AI's.. As for NPC's remembering things, that isn't new.. It's called faction, and in many games in the past you could have neutral faction and get no help from NPC's in a fight, but if you were allied for example and in danger, a NPC would come to your aid if you were in range.. That is not rocket science.. If you keep track of every groups faction, which original EQ had many, it's not that hard to have a deep AI similar to Storybricks..
ALL mobs in any game have 2 choices.. They are given random orders, or they follow a script of yes/no code checks.. This applies to combat as well.. Either the mob will randomly attack anyone in a given range, or it will follow code instructions.. So devs can spew all the BS they want about smart AI actions, but once that code is discovered, all their effort is meaningless then.. Players will find out what the code is and abuse it and use it..... So removing taunt does nothing in the end.. UNLESS there are more random mobs then they are leading on.. And if that is the case, then it's not smart AI, it's Random AI and we have Zergfest..
You dolts realize that the whole MIT thing was just sarcasm right?
Storybricks is script, it just has more "yes and no" checks then dumb AI's.. As for NPC's remembering things, that isn't new.. It's called faction, and in many games in the past you could have neutral faction and get no help from NPC's in a fight, but if you were allied for example and in danger, a NPC would come to your aid if you were in range.. That is not rocket science.. If you keep track of every groups faction, which original EQ had many, it's not that hard to have a deep AI similar to Storybricks..
ALL mobs in any game have 2 choices.. They are given random orders, or they follow a script of yes/no code checks.. This applies to combat as well.. Either the mob will randomly attack anyone in a given range, or it will follow code instructions.. So devs can spew all the BS they want about smart AI actions, but once that code is discovered, all their effort is meaningless then.. Players will find out what the code is and abuse it and use it..... So removing taunt does nothing in the end.. UNLESS there are more random mobs then they are leading on.. And if that is the case, then it's not smart AI, it's Random AI and we have Zergfest..
I was following StoryBricks back when they were trying to make a game out of it. It's not random choices, it's a decision tree kind of thing. It's not all that different from how people decide to do things.
Given a set of "wants", the AI will follow the decision tree to achieve the goal. The example they use is a group of gnolls or something wanting a relatively quiet area, where a few travelers come through. If an area gets to busy, the gnolls will search out another area. If there aren't enough travelers to attack, they'll search out another area. By itself, that one set of gnolls isn't very intelligent.
However, the NPC guards use the same system. So players can report a gnoll sighting to the guards, and the guards may search out and drive the gnolls off, or they may offer a reward to players for killing the gnolls. Now the system is a little more intelligent and responsive to the players.
None of the NPCs will do something that the developers don't first supply an action for. The gnolls will not start writing poetry, but they may try to bribe players into leaving them alone, or they may hire assassins to kill a particularly aggressive player. The AI is more intelligent than current MMORPG AI systems, and it's certainly more responsive than existing AI systems.
The combat stuff is pretty trivial in comparison. We've had advanced combat AI for years now. We could have combat AI that wipes the floor with players by always making the right choices, and that includes group combat. Blizzard could do that in WoW. They need to be able to do stuff like that for testing purposes. They just don't turn that AI on the players because it wouldn't be fun and would be a road block towards the goal of dropping loot for the players.
I can not remember winning or losing a single debate on the internet.
You are expecting too much from the AI that spawns mobs
I agree. This has already existed, which is why it is PR hype. It isn't hard to make a computer better than players, but it would be stupid to turn it on, because no one wants to play a near impossible game.
Those aren't my expectations, those are the descriptions of the AI behavior from the StoryBricks developers and from SOE. They were working on the StoryBricks AI years before SOE ever started talking to them. It was originally a system for building procedural story content in a story based game world. That's just how the AI works. It doesn't just "spawn mobs".
I can not remember winning or losing a single debate on the internet.
Yeah, I'm gonna have to pretty much disagree with most of your post here. First of all the trinity is not just tank/dps/healer. There is also cc. Like has been pointed out the original trinity was tank/dps/cc. The term trinity is now used generically for a system that has defined roles. It also doesn't "force" people into 3 roles. There is healing, dps, tanking, offtanking, CC, buffing and hybrids which can be a combination of roles but still unique in their own right in what they offer to a group. For example EQ had bards which didn't neatly fit into a category but were damn useful.
Only warriors could tank bosses? SK's were useless? Sorry but that is just wrong. We had a SK main tank in EQ that tanked the bulk of our raids and was one of the better tanks I have seen. Sure, some fights we used a warrior or a paladin but that was the beauty of the system. There were many possible strategies and the class system allowed for creative solutions and the ability to be successful with what you had rather than using the same strat/raid makeup that every other guild used. We also had some really great paladin tanks. We experimented with different things. Sometimes it worked and other times it didn't. That was the fun and was especially gratifying when we came up with our own method of doing something.
Classes in EQ weren't balanced to all be exactly equal because they had actual variety. Yes, clerics were the best healers but you know what? Our guild used druids on CH rotations in a pinch. Sure the 75% CH was not as good but stick a druid in between 2 clerics in the rotation and it worked. Not ideal but it worked. Druids weren't thought of as inferior clerics because they brought DPS and buffs and snares and roots that the cleric didn't have. This is true flexibility and variety. Not making everyone's abilities the same but with different animations/weapons/gear to give the illusion that you are doing something different.
WoW this thread is still alive and ongoing.. as much as i know i wrote something similar to lizard at page 5.. and with that the thread should basicly end.
@lizard: I do agree.
And i want to add to your last paragraph ( and i guess i have even wrote that somewhere in this thread, but can't be bothered to read all 30 pages yet again).
The target of storybrick in combat is not to be a perfect AI, to make always the right dicisions, it is much more to make reasonable and sometimes even wrong(but reasonable) decisions, like a lot of humans do it every day.
That a group of mobs don't act like one men attacking the perfect target, but much more behave like a group, with every single one of them behave a little bit different, a littly bit asynchron, and do even other decisions. With other words, they try to give the mobs the ability to make mistakes.. and not either the perfect counter, or just one stupid move all of them do.
And storybrick is by the way not the first AI tool, which tried it.. anyone interested can look up about emotional AI and you can find some interesting read. (although it was never used in a MMO, at least as much as i know)
Upgrading AI with something like storybricks is a good decision all around, in theory.
The problem many people have is upgrading AI in such a way that it removes traditional role based combat, and the false notions that its 1) the only way to improve AI, 2) that its somehow the more logical alternative and/or 3) the logical evolution of game AI. These notions are an insult to the intelligence of players who have any understanding of game mechanics and artificial intelligence on even the most basic level. SOE's sudden hostility and outright mockery of the previous forms of video game AI (especially mmorpg AI), specifically threat management (ie. "the trinity"), has been nothing less than ridiculous and a shameless attempt to promote a new form of combat gameplay, not just advanced AI as they'd have you believe. The scary thing is, most of their staff are apparently drinking their own kool-aide, which became obvious during SOE live when no less than a dozen times I heard them refer to threat management and hate generating mechanics as something along the lines of "insulting a mobs mother." That is not a good sign, and I'm slightly embarrassed on their behalf.
It comes down to a matter of gameplay preference, not accepting the logical evolution of AI. Many MMORPG players, especially those who have enjoyed EQ, would prefer to see them build upon the foundation of threat management combat which promotes the established roles found in traditional rpgs. Some of which are more action based (tanks and some melee) and others which are more layed back or passive. This traditionally has allowed for all types of people to participate and enjoy mmorpgs, both those who enjoy being the leader with better situational awareness, aim and reflexes, as well as those looking for a role with less responsibility, less twitch capabilities, and frankly a role that isn't stressful (some people work for a living). SOE quite obviously intends to leave these combat mechanics behind altogether in favor of arcade action-adventure combat where all the "roles" (if you can call them that) are action based and, at one time or another, become the de factor "tank" by adapting to combat scenarios and utilizing movement abilities like jumps, dodges, rolls or teleports. This combat has proven fun in some games, but has always been zergy in massively MULTIPLAYER online rpgs, hence the resistance from the more experienced, less naive mmo player.
I, for one, am not fool by this, nor am I subverted by their juvenile attempts to mock the AI that has served games well for decades (and will continue to serve), regardless of how trivial it has become in modern MMOs. The triviality of modern MMOs should be attributed not to the code, but the philosophies industry leaders have employed to generate a larger playerbase and make their game more all-inclusive ($$$). Fortunately for SOE, the vast majority of their target audience hasn't the slightest concept of how mmos work, or should work, so they stand to attract quite a playerbase with their grand claims and hyperbolic remarks.
I, however, will not be among them.
Wow... a triple straw man...straw men rather. Nice to create your own false notions and then shoot them down...
1. "The only way to improve AI"... who, other than you is saying that there is only one way forward?
2. "it's somehow the more logical alternative" ... it's an alternative...it's different... more logical? Not sure that's the point. More like a player would behave in that situation? Yeah, that's a bit closer.
3. "The logical evolution of game AI" ... there you go with that "logical" thing again.
What in fact is happening here, and in other places--EQN is not by any means the only MMORPG in development attempting new mob combat behavior--is a recognition that things have been done in one specific way for a long time and that doing them in a different way might be more fun.
So they're trying a new way. And this new way involves attempting to have mobs and friendly NPCs behave in a manner that is more realistic... more like how they would behave if there was a player controlling them instead of just a canned script.
I won't bore you with the technical details--I've already linked those in previous posts in this thread. Suffice it to say, that this system also has tweakable difficulty and the people designing it are fully aware that the bottom line is enjoyment and satisfaction.
This new way will be different than the traditional trinity-based combat with almighty tank aggro kludges. Whether it will be more enjoyable remains to be seen.
But I for one am looking forward to seeing this new way of PVE combat. God knows that when MMOs became big business, innovation and risk taking took a back seat to the tried and true financially successful formulas. It's nice to see a team with the cojones to try to do something different. That alone makes me inclined to cut them a lot of slack and just wait and see what they come-up with.
“Microtransactions? In a single player role-playing game? Are you nuts?”
― CD PROJEKT RED
C:\Users\FF\Desktop\spin move.gif
"Coming from the software development world I know for a FACT that SOE is not employing people from MIT's AI department for their game. This would be the only way to produce AI even close to challenging enough to make the "Trinity" system obsolete. However, even MIT hasn't created AI advanced enough to make this claim." - Bearknight
Well, we either have someone who has a very unique take on the OP's message, or bearknight just outed his alt.
There isn't a "right" or "wrong" way to play, if you want to use a screwdriver to put nails into wood, have at it, simply don't complain when the guy next to you with the hammer is doing it much better and easier. - Allein
"Graphics are often supplied by Engines that (some) MMORPG's are built in" - Spuffyre
If anything, the 4th pillar was viewed here with a decent amount of skepticism. On the AI, it's an alternative to the most common form of combat and mob AI. If MMOs adopt it, that's a pretty significant change, primarily because most MMOs are centered around combat.
There isn't a "right" or "wrong" way to play, if you want to use a screwdriver to put nails into wood, have at it, simply don't complain when the guy next to you with the hammer is doing it much better and easier. - Allein
"Graphics are often supplied by Engines that (some) MMORPG's are built in" - Spuffyre
And yet here you are, repeatedly biting and unfortunately making everyone here suffer your constant complaints.
SOE accumulated quite some knowledge about AI. Vanguard has advanced AI. They just switched it off a few weeks after release. Back then mobs ran away and came back with friends for example. EQ2 has some quite good mercanaries.
Both games had to tune down AI because of performance problems. The EQN engine is better performing, allowing the CPU to compute more AI.
They had to tune it down because people don't really want hard fights, they want it to be predictable.
Mobs ran in eq as well ... in 1999. It's not a matter hardware performance, most player now could not accept a real challenge.
+1 The AI is out there, it just can't be used as it wold 'win' every time.
@OP I understand your frustration, but the points you bring up are largely half truths and inaccurate. Sadly, your points on the trinity are accurate and I hope someone makes it so a balanced group is required to complete group content.
If you want a new idea, go read an old book.
In order to be insulted, I must first value your opinion.