The problem with mmorpgs and story...
by Jusitn Scifres · in Game Design and Creative Issues · 06/14/2006 (12:43 pm) · 115 replies
This is a bit of a philiposhpical / literary venture so bear with me... there's an idea in here, i promise...
the reason why mmorpgs are hard to become motivated with is that there is never a sense of story. sure, there's a main quest, but you can't help but to realize that thousands of others have gone before you, and that the quest really means nothing in the longterm. In fact, you're probably PLing with a guy who's done it like fifty million times. I beat the boss. whoopee.
the reason that this happens, is because from a literal standpoint, there is no individuality between characters. essentially, people are just sitting in on the games of others. your character has beat the evil ice wizard. his has not. wtf? how many fricking ice-wizards are there? the only way to remedy this problem is to make a singular, cohesive, and continuing story that all players experience simultaneously.
umm... okay. let's pretend that was easy for a moment. so new players are to miss out on 30 or so in-game years of story?
well... yeah.
part of the reason that the games aren't engaging is that the characters themselves never really feel that they are the movers and shakers of the game world. this comes partly from the arbitrary multi-universe storylines and level caps. most players (or at least me) lose interest in mmorpgs because they feel like just another face in the crowd.
remember in galaxies when there was finally a jedi? that game got fun in a hurry. suddenly, there was a player-motivated series of events going down. people became engaged. the game became a little bit more of a real tangible world.
okay, let me give an example game that would cover my philosophy. excuse the cheese, writing off the cuff:
the reason why mmorpgs are hard to become motivated with is that there is never a sense of story. sure, there's a main quest, but you can't help but to realize that thousands of others have gone before you, and that the quest really means nothing in the longterm. In fact, you're probably PLing with a guy who's done it like fifty million times. I beat the boss. whoopee.
the reason that this happens, is because from a literal standpoint, there is no individuality between characters. essentially, people are just sitting in on the games of others. your character has beat the evil ice wizard. his has not. wtf? how many fricking ice-wizards are there? the only way to remedy this problem is to make a singular, cohesive, and continuing story that all players experience simultaneously.
umm... okay. let's pretend that was easy for a moment. so new players are to miss out on 30 or so in-game years of story?
well... yeah.
part of the reason that the games aren't engaging is that the characters themselves never really feel that they are the movers and shakers of the game world. this comes partly from the arbitrary multi-universe storylines and level caps. most players (or at least me) lose interest in mmorpgs because they feel like just another face in the crowd.
remember in galaxies when there was finally a jedi? that game got fun in a hurry. suddenly, there was a player-motivated series of events going down. people became engaged. the game became a little bit more of a real tangible world.
okay, let me give an example game that would cover my philosophy. excuse the cheese, writing off the cuff:
About the author
#22
shhh, don't tell anyone ;)
I have to STRONGLY dissagree with your PD ideas Alexander. Here is why: First, what you describe isn't even PD. You say you keep a soul that has your basic stuff like xp, and then just keep takin over new bodies? What makes PD actually permanent? The fact that you LOOSE all that xp makes it PD. The fact that any items, money, character, etc are all lost. THAT is perma-death. What you describe is just another way to handle death without having PD.
This is completely wrong again. Players DO get pissed when they die in counter-strike, bf1942, etc. They don't care as much as they would if they lost an MMO character, but they still care. MMO's are different in so many ways it's not even a good analogy. When I play CS, I pick from 4 characters total. Half the team has the same skin as me and looks just like me which means I'm disconnected from my character. THAT is why I don't care as much. I don't care if I loose this character because I don't really loose anything. I don't even loose the MONEY I had before death (in CS). All I loose is the ability to finish that round (in cs) or I have to start back over at a spawn point (BF2).
In the MMO world, PD means I loose any skills I ever got, any items I ever found or bought, my character that I spent time customising, all my experience points, my level.... The list goes on and on. PD in an MMO would ONLY work on a PvP based MMO or in PvP Zones on every other MMO. Where people are CONSTANTLY there to do battle and nothing but battle. Even then, I think it wouldn't work out. No one wants to build a bad ass character only to find a slightly better one that can jack you and then you have to start over.
When building an MMO you have to weigh the realism with the fun aspects. Realism should NEVER be more important that fun. Even White Wolf themselves (who I hold as the greatest role playing publishers/writers of all time) say that 'these are just a set of rules, don't let them interfere with the fun of the game'. Key word: "GAME". This isn't real life, don't fall into the 'realism' trap and try to realistically drive your player base into the drain.
06/21/2006 (7:24 am)
Quote:Instead a "next gen" mmo should use a modern experience system, like White Wolf's
shhh, don't tell anyone ;)
I have to STRONGLY dissagree with your PD ideas Alexander. Here is why: First, what you describe isn't even PD. You say you keep a soul that has your basic stuff like xp, and then just keep takin over new bodies? What makes PD actually permanent? The fact that you LOOSE all that xp makes it PD. The fact that any items, money, character, etc are all lost. THAT is perma-death. What you describe is just another way to handle death without having PD.
Quote:Well, the second argument does not apply to PD settings. Think about any multiplayer shooter (battlefield, ..). Your character always dies after a few minutes. Big deal. You play with another char. Players don't grieve over the deaths of their chars.
This would be similar in a PD mmo game. People know from the beginning that if their character dies, they have to play another one (99% of deaths). A heroic death would result in tales told long after, which means true role playing.
This is completely wrong again. Players DO get pissed when they die in counter-strike, bf1942, etc. They don't care as much as they would if they lost an MMO character, but they still care. MMO's are different in so many ways it's not even a good analogy. When I play CS, I pick from 4 characters total. Half the team has the same skin as me and looks just like me which means I'm disconnected from my character. THAT is why I don't care as much. I don't care if I loose this character because I don't really loose anything. I don't even loose the MONEY I had before death (in CS). All I loose is the ability to finish that round (in cs) or I have to start back over at a spawn point (BF2).
In the MMO world, PD means I loose any skills I ever got, any items I ever found or bought, my character that I spent time customising, all my experience points, my level.... The list goes on and on. PD in an MMO would ONLY work on a PvP based MMO or in PvP Zones on every other MMO. Where people are CONSTANTLY there to do battle and nothing but battle. Even then, I think it wouldn't work out. No one wants to build a bad ass character only to find a slightly better one that can jack you and then you have to start over.
When building an MMO you have to weigh the realism with the fun aspects. Realism should NEVER be more important that fun. Even White Wolf themselves (who I hold as the greatest role playing publishers/writers of all time) say that 'these are just a set of rules, don't let them interfere with the fun of the game'. Key word: "GAME". This isn't real life, don't fall into the 'realism' trap and try to realistically drive your player base into the drain.
#23
I'm finding that that same DB setup that I put into place for socketed items seems to lend itself to DB generated story arcs that can be pieced together to form larger stories. That concept needs some work (work I've already done), but if it plays out well, then you could conceivably generate storylines of various complexities in realtime.
Another thing people should be aware of is that the AI in MMO's will need a serious upgrade to bring about a new storyline system, especially one that is dynamic. Now, NPC's just have a few scripted missions that they give out to players, but they don't really interact with you. In some cases, this is necessary, as the NPC is "serving" many players, and so cannot do much more than stand there and talk to two or three people at a time. But, other NPC's can be generated dynamically to serve content to a single user, and that NPC can be "put away" when that user is gone, stored for the next time the user comes around. Or you can use a memory system for some NPC's so that they can keep track of "memorable" events that will tie them to certain players. More importantly, the NPC should have access to all of the controls and skills that a player has, if you want true flexibility in it's AI.
06/21/2006 (7:54 am)
Hence, PD should be something voluntary, that is rewarded if it actually occurs. But we're still OT, as the thread was about stories in MMO's, and there is very little discussion on that point.I'm finding that that same DB setup that I put into place for socketed items seems to lend itself to DB generated story arcs that can be pieced together to form larger stories. That concept needs some work (work I've already done), but if it plays out well, then you could conceivably generate storylines of various complexities in realtime.
Another thing people should be aware of is that the AI in MMO's will need a serious upgrade to bring about a new storyline system, especially one that is dynamic. Now, NPC's just have a few scripted missions that they give out to players, but they don't really interact with you. In some cases, this is necessary, as the NPC is "serving" many players, and so cannot do much more than stand there and talk to two or three people at a time. But, other NPC's can be generated dynamically to serve content to a single user, and that NPC can be "put away" when that user is gone, stored for the next time the user comes around. Or you can use a memory system for some NPC's so that they can keep track of "memorable" events that will tie them to certain players. More importantly, the NPC should have access to all of the controls and skills that a player has, if you want true flexibility in it's AI.
#24
On death:
I really do believe that the 'galaxies' style (i really don't know since i've never played) of saving your character for money then starting there when you die is the way to go. all i want is for people to be afraid enough of death that they run away from a dangerous situation. i want cowardice. because heros are not people with no fear, but those who face it. Any over-complication of that system only adds to death's novelty. and, it's probably important to mention that there really wouldn't be any PvP in this story, only players shoulder to shoulder against Fallen. plus, there would be ways that players could raise eachother on the field, and it would take a very long time to 'die'. after a battle there would probably be sweeper teams of medics going around... my main concept is that teamwork takes the place of 'go here, kill this'. so it would be pretty hard to die. and once it happened, it wouldn't be so bad, just extremely inconvienient. But that wouldn't be it. I'm trying to foster a mood of cooperation here. that when you die, there's people there to help. (btw, pvp would still happen, but only in the form of simulations. it would be possible to bet items and even gain stats through simulation, but no one would get killed.)
On quests:
I never really like the game to hold my hand on a quest. I think some of the most fun i ever had doing quests was in morrowind.
'here's this.'
'well, what do i do with it?'
'fuck if i know, kid.'
but i'm thinking even less structure than that. i want people to have to go and make their own observations and accomplishments. I want people to wander around, find some thing, and then realize later that there's a big group of people who are willing to kill for it. at some level, there has to be totally unique things in the game that the overwhelming majority of people will only hear about. that's the only way. Not to even have quests.
On npc's:
fuck 'em. the players should be the npcs. no quests, no npcs required. want to buy something? buy it from a character. items fixed? characters can do it. not only will this actually make a real economy, but you could put a set amount of money in the world and not have some random creature drop it. 'fearsome bunny dropped 12 gold!'
on story:
as said before, the story, if you want it to be truly character motivated, has to run only on greed, fear, and general nastiness. because, after all, that's what drives real stories. if everyone was holding hands and singing, it wouldn't make for a very fun time. possible idea: have gods. each server would have gods. most likely, people who made the game, who just pull a parthenon and screw with mortals. have feuds using mortals as pawns. you would have to have a set of rules, but hey, that would be pretty fun i think. there's a whole lot of possibilities there...
or, have a global threat or pursuit. something that everybody wants to contribute on.
either way, something needs to change.
06/21/2006 (8:48 am)
Alrighty. valid points all. opinion time.On death:
I really do believe that the 'galaxies' style (i really don't know since i've never played) of saving your character for money then starting there when you die is the way to go. all i want is for people to be afraid enough of death that they run away from a dangerous situation. i want cowardice. because heros are not people with no fear, but those who face it. Any over-complication of that system only adds to death's novelty. and, it's probably important to mention that there really wouldn't be any PvP in this story, only players shoulder to shoulder against Fallen. plus, there would be ways that players could raise eachother on the field, and it would take a very long time to 'die'. after a battle there would probably be sweeper teams of medics going around... my main concept is that teamwork takes the place of 'go here, kill this'. so it would be pretty hard to die. and once it happened, it wouldn't be so bad, just extremely inconvienient. But that wouldn't be it. I'm trying to foster a mood of cooperation here. that when you die, there's people there to help. (btw, pvp would still happen, but only in the form of simulations. it would be possible to bet items and even gain stats through simulation, but no one would get killed.)
On quests:
I never really like the game to hold my hand on a quest. I think some of the most fun i ever had doing quests was in morrowind.
'here's this.'
'well, what do i do with it?'
'fuck if i know, kid.'
but i'm thinking even less structure than that. i want people to have to go and make their own observations and accomplishments. I want people to wander around, find some thing, and then realize later that there's a big group of people who are willing to kill for it. at some level, there has to be totally unique things in the game that the overwhelming majority of people will only hear about. that's the only way. Not to even have quests.
On npc's:
fuck 'em. the players should be the npcs. no quests, no npcs required. want to buy something? buy it from a character. items fixed? characters can do it. not only will this actually make a real economy, but you could put a set amount of money in the world and not have some random creature drop it. 'fearsome bunny dropped 12 gold!'
on story:
as said before, the story, if you want it to be truly character motivated, has to run only on greed, fear, and general nastiness. because, after all, that's what drives real stories. if everyone was holding hands and singing, it wouldn't make for a very fun time. possible idea: have gods. each server would have gods. most likely, people who made the game, who just pull a parthenon and screw with mortals. have feuds using mortals as pawns. you would have to have a set of rules, but hey, that would be pretty fun i think. there's a whole lot of possibilities there...
or, have a global threat or pursuit. something that everybody wants to contribute on.
either way, something needs to change.
#25
if your in lets say WoW you get a quest like
*Walk to there and kill 20 of those come back and get some reward*
there is not a story behind it you don't feel the desire i whant to know how the story goes on and complete it with the hope on getting the ansfer
if you where to make all the quests a one timer and to put a complete story behind all the quests and greater rewards then you would get a dynamic story if you where to continue the quests with a kind of editor where you could make quests based on some events or posibilety's where you have knowlage of at that time and give it to anothere player
And for the npc's:
i agree on the part of the bunny's dropping gold and items that are bigger then themselfs
but for a starting point for instance a camp with guards its vital so newcomers won't get killed in battles nearby
and its ecential to have npc's to create the starting story and to help you learn aboud it
at least thats my opinion
06/21/2006 (8:59 am)
Jusitn on the quests i think the fun of quests is in the big story behind itif your in lets say WoW you get a quest like
*Walk to there and kill 20 of those come back and get some reward*
there is not a story behind it you don't feel the desire i whant to know how the story goes on and complete it with the hope on getting the ansfer
if you where to make all the quests a one timer and to put a complete story behind all the quests and greater rewards then you would get a dynamic story if you where to continue the quests with a kind of editor where you could make quests based on some events or posibilety's where you have knowlage of at that time and give it to anothere player
And for the npc's:
i agree on the part of the bunny's dropping gold and items that are bigger then themselfs
but for a starting point for instance a camp with guards its vital so newcomers won't get killed in battles nearby
and its ecential to have npc's to create the starting story and to help you learn aboud it
at least thats my opinion
#26
Interesting concept, but you aren't planning to make the players wait for these teams to find their bodies for them to rez? That would be a turn-off to a lot of people.
Actually, real stories are driven by conflict, and not just negative conflicts, but conflicts of a less negative sort. If you only used stories that had the motivating factors of greed, fear, and nastiness, then it would be a dismal feeling game. I remember Anarchy Online, when I played it, had missions that were supposedly dynamic, but all basically said the same thing "go here and kill copious amounts of MOB's then come back for your reward", and after a short period of time, that's all it was, despite their efforts of giving back story and events, etc. There has to be a lighter side of the equation in order for it to balance out, or else it will seem bland to the player after a while, and I think this is one of the things that WoW has going for it: Even though the quests are scripted and get old after your second or third character, they don't take themselves too seriously, and incorporate humor into some of their quests to lighten the mood. Making the player laugh at times goes a long way.
I also agree with bunnies dropping gold, but I disagree with NPC's, as there is no guarantee that your world will be able to run in the begining with only players, and the effects that that has can skew the experience for everyone down the road. NPC's are a very valuable part of online games, and add a lot to the story and ambience of the world. However, listening to your design, it sounds more like a persistent-character online FPS than an MMO, as it's mostly combat based with stats and character/world persistence, and less of an emphasis on role-playing. Not that that's a bad thing (especially if the game turns out to be fun), but that being the case, then you'd be fine with not needing NPC's or overly complex stories/missions.
06/21/2006 (1:26 pm)
Quote:there would be ways that players could raise eachother on the field, and it would take a very long time to 'die'. after a battle there would probably be sweeper teams of medics going around...
Interesting concept, but you aren't planning to make the players wait for these teams to find their bodies for them to rez? That would be a turn-off to a lot of people.
Quote:as said before, the story, if you want it to be truly character motivated, has to run only on greed, fear, and general nastiness. because, after all, that's what drives real stories.
Actually, real stories are driven by conflict, and not just negative conflicts, but conflicts of a less negative sort. If you only used stories that had the motivating factors of greed, fear, and nastiness, then it would be a dismal feeling game. I remember Anarchy Online, when I played it, had missions that were supposedly dynamic, but all basically said the same thing "go here and kill copious amounts of MOB's then come back for your reward", and after a short period of time, that's all it was, despite their efforts of giving back story and events, etc. There has to be a lighter side of the equation in order for it to balance out, or else it will seem bland to the player after a while, and I think this is one of the things that WoW has going for it: Even though the quests are scripted and get old after your second or third character, they don't take themselves too seriously, and incorporate humor into some of their quests to lighten the mood. Making the player laugh at times goes a long way.
I also agree with bunnies dropping gold, but I disagree with NPC's, as there is no guarantee that your world will be able to run in the begining with only players, and the effects that that has can skew the experience for everyone down the road. NPC's are a very valuable part of online games, and add a lot to the story and ambience of the world. However, listening to your design, it sounds more like a persistent-character online FPS than an MMO, as it's mostly combat based with stats and character/world persistence, and less of an emphasis on role-playing. Not that that's a bad thing (especially if the game turns out to be fun), but that being the case, then you'd be fine with not needing NPC's or overly complex stories/missions.
#27
- Nokill -
the 'quests' themselves won't really be thigns that one person can ever accomplish by themselves. in fact, the quests won't even be required by the game world. But if completed, they will reveal a larger story to the entire server (and hopefully, servers). As the game server's objectives are completed, the story will pan out. how, when, and why will be left entirely up to the players, and the leadership therein. it won't be just go here kill this. it'll be scout ahead, report back, go here, kill this, find place, fix place, defend place, guard place; for example. and in the process, let's say some vital information about the fallen is discovered. this is just a generalization, but it gives you the basic idea. but a quest like this requires all sorts of characters at varying skill levels. raw manpower is always needed to defend an area. infantry is a must.
as far as a 'quest editor', no need, since another player could just tell you what to do. it would be free-form enough to allow that.
as far as starting places, you would start in a well-fortified citiy. not impregnable, but reliable enough to shelter the newbs.
i really have a lot of derision concerning npc's. i don't like them at all. they just stand there 24 hours a day and say the same thing to everyone. a story could be much more realistically conveyed in the form of modern media, word of mouth, or server-wide emails updating everyone on news from their server and others. and background information could be gained through simply talking to the old-timers.
- ted -
the medic teams i was referring to are just a strategy i forsee some players employing. it would serve the dual purpose of healing the wounded and increasing those ever important medic skills. the medics would be actual players that come after the battle (or during) and just start rezing people. like i said, just an eventuality i see...
conflict by definition is the opposition of two misaligned forces. assuming that the 'good' force is motivated by good emotions, the aggressor 'evil' force is negative. name a good story that did not conatin misery and evil. they don't exist. you don't read good books with two protagonists. especially in the gritty sort of world i want to create. why can't i make a depressing hole-in-the ground ambience to the game? war sucks. nobody should be holding hands and singing kumbayah in that world. looking out for number one is a fucked up attitude but unfortunately, it happens. but hey, since when is different bad? I don't think i've been depressed by the atmosphere of a game for a while. poe, anyone?
nah... i really don't see the persistent first person thing going down. i would much rather see this go down like final fantasy XI... there was a feeling to that game that was a little unique... i'd played many mmorpgs and never got the same vibe. I'm not saying it's the best mmorpg (far from it), just that it had this unique voice and feel that i dug on. that's how i would do an mmo. i like semi-turn based. the strategy, the thinking. i'm not a real big fan of micro-managing aggro, though... but that's definately off-topic.
point is, i think that history, purpose, and quests (as well as free-form dungeon crawling) can all be accomplished in a non-handholding environment. the players that are dedicated and intelligent would naturally rise to the top, motivating the story. as for the grunts, they would have just as important roles. there are no armies of specialists. players would communicate naturally, as players do, to pass the story on to eachother. plus, the media forms detailed above.
there's the two cents, anyway.
06/22/2006 (2:01 pm)
Responses:- Nokill -
the 'quests' themselves won't really be thigns that one person can ever accomplish by themselves. in fact, the quests won't even be required by the game world. But if completed, they will reveal a larger story to the entire server (and hopefully, servers). As the game server's objectives are completed, the story will pan out. how, when, and why will be left entirely up to the players, and the leadership therein. it won't be just go here kill this. it'll be scout ahead, report back, go here, kill this, find place, fix place, defend place, guard place; for example. and in the process, let's say some vital information about the fallen is discovered. this is just a generalization, but it gives you the basic idea. but a quest like this requires all sorts of characters at varying skill levels. raw manpower is always needed to defend an area. infantry is a must.
as far as a 'quest editor', no need, since another player could just tell you what to do. it would be free-form enough to allow that.
as far as starting places, you would start in a well-fortified citiy. not impregnable, but reliable enough to shelter the newbs.
i really have a lot of derision concerning npc's. i don't like them at all. they just stand there 24 hours a day and say the same thing to everyone. a story could be much more realistically conveyed in the form of modern media, word of mouth, or server-wide emails updating everyone on news from their server and others. and background information could be gained through simply talking to the old-timers.
- ted -
the medic teams i was referring to are just a strategy i forsee some players employing. it would serve the dual purpose of healing the wounded and increasing those ever important medic skills. the medics would be actual players that come after the battle (or during) and just start rezing people. like i said, just an eventuality i see...
conflict by definition is the opposition of two misaligned forces. assuming that the 'good' force is motivated by good emotions, the aggressor 'evil' force is negative. name a good story that did not conatin misery and evil. they don't exist. you don't read good books with two protagonists. especially in the gritty sort of world i want to create. why can't i make a depressing hole-in-the ground ambience to the game? war sucks. nobody should be holding hands and singing kumbayah in that world. looking out for number one is a fucked up attitude but unfortunately, it happens. but hey, since when is different bad? I don't think i've been depressed by the atmosphere of a game for a while. poe, anyone?
nah... i really don't see the persistent first person thing going down. i would much rather see this go down like final fantasy XI... there was a feeling to that game that was a little unique... i'd played many mmorpgs and never got the same vibe. I'm not saying it's the best mmorpg (far from it), just that it had this unique voice and feel that i dug on. that's how i would do an mmo. i like semi-turn based. the strategy, the thinking. i'm not a real big fan of micro-managing aggro, though... but that's definately off-topic.
point is, i think that history, purpose, and quests (as well as free-form dungeon crawling) can all be accomplished in a non-handholding environment. the players that are dedicated and intelligent would naturally rise to the top, motivating the story. as for the grunts, they would have just as important roles. there are no armies of specialists. players would communicate naturally, as players do, to pass the story on to eachother. plus, the media forms detailed above.
there's the two cents, anyway.
#28
Try Puzzle Pirates ( http://www.puzzlepirates.com/ ). When at sea with a vessel if you get attacked and lose you lose booty, and can even get to the point you are sluggishly just trying to get back to land because your ship is so damaged. Players aren't terrified of losing but they certainly think twice if they don't think they will, some even to the extent they will evade battle.
Another example is when a war actually occurs. Ships can be sunk, and trust me that's a big deal as it takes _alot_ to get a ship and any pirates on it have a chance to lose limbs. Pirates can end up with peg-legs, hook arms, lose an eye, etc. The character isn't gone but it shows that you lost, but if really determined they can cure their injuries by getting a potion and in time they should have enough money to get their ship back.
06/22/2006 (2:13 pm)
Sorry didn't read all of this discussion yet but saw the note on making dying rough or scary.Try Puzzle Pirates ( http://www.puzzlepirates.com/ ). When at sea with a vessel if you get attacked and lose you lose booty, and can even get to the point you are sluggishly just trying to get back to land because your ship is so damaged. Players aren't terrified of losing but they certainly think twice if they don't think they will, some even to the extent they will evade battle.
Another example is when a war actually occurs. Ships can be sunk, and trust me that's a big deal as it takes _alot_ to get a ship and any pirates on it have a chance to lose limbs. Pirates can end up with peg-legs, hook arms, lose an eye, etc. The character isn't gone but it shows that you lost, but if really determined they can cure their injuries by getting a potion and in time they should have enough money to get their ship back.
#29
I used to play "Galaxies" and their death system wasn't scary at all. It was a money sink. Money was readily available and if you didn't have the money you would still clone just with wounds and damage to your equipment which for the most part could be repaired and/or replaced. An interesting system but no where close to Permadeath.
NOT to say that permadeath DIDN'T exist in SWG. The Jedi had it as a penalty for a short while but no one liked it so it was dropped for loss of XP. It can be argued how well this worked for SWG (and it often is)- but IMO it is a great way to scare a certain type of player away from dying for dying's sake. I say "certain type of player" because this really only affects those that are playing the game to build their character. Which is a majority of the RPG player base. This system brought about some VERY intense emotion between the Jedi- (who lost the XP) and the Bounty Hunters (who killed them causing the loss of XP).
In the game I am making I intend to work a dynamic similar to this- but a little more involved. I still think Permadeath is an option and doesn't have to be this black and white "you die and lose all your stuff" penalty. When one door closes another door opens, no?
Your game idea sounds really good. Very interesting. The biggest problem that I see is that you seem to put more trust in your playerbase than I EVER would. You expect them to come up with their own quests and to have to group to complete all the missions- hell, you don't even want to have NPCs. This could be a big problem.
In the case of your playerbase working coming up with their own missions, interdependancies, grouping, etc. A lot of players are casual players- who only play for a couple of hours each day at most and don't want to spend that time searching for people to help them complete a mission.
Many, many more players are soloists. People who enjoy grouping but for the most part want to adventure on their own. And though MMORPGS are multiplayer games- video games are solo by nature. So while a majority of the playerbase likes the IDEA that other live beings are running around and they can play with their friends they HATE the fact that they always have to be in a group to accomplish anything.
Lastly, I have to disagree with your thoughts on NPCs. In fact I think the exact opposite. Everyone wants a "virtual world" in these MMOs but then only program NPCs that do nothing until players engage them. I say NPCs should always be shaping and changing the world- Especially in a game that has a war going on. Cities should be rising and falling- crimes and saves should be happening- NPCs should have a HUGE relationship with each other and probably out number the playerbase in great numbers.
Create a world that works without players- then let the players come in and play in it.
That is the goal for my game.
my two cents.
06/24/2006 (12:24 pm)
Alright I'm gonna drop a dime or two worth of info in here cuase it's very interesting. I used to play "Galaxies" and their death system wasn't scary at all. It was a money sink. Money was readily available and if you didn't have the money you would still clone just with wounds and damage to your equipment which for the most part could be repaired and/or replaced. An interesting system but no where close to Permadeath.
NOT to say that permadeath DIDN'T exist in SWG. The Jedi had it as a penalty for a short while but no one liked it so it was dropped for loss of XP. It can be argued how well this worked for SWG (and it often is)- but IMO it is a great way to scare a certain type of player away from dying for dying's sake. I say "certain type of player" because this really only affects those that are playing the game to build their character. Which is a majority of the RPG player base. This system brought about some VERY intense emotion between the Jedi- (who lost the XP) and the Bounty Hunters (who killed them causing the loss of XP).
In the game I am making I intend to work a dynamic similar to this- but a little more involved. I still think Permadeath is an option and doesn't have to be this black and white "you die and lose all your stuff" penalty. When one door closes another door opens, no?
Your game idea sounds really good. Very interesting. The biggest problem that I see is that you seem to put more trust in your playerbase than I EVER would. You expect them to come up with their own quests and to have to group to complete all the missions- hell, you don't even want to have NPCs. This could be a big problem.
In the case of your playerbase working coming up with their own missions, interdependancies, grouping, etc. A lot of players are casual players- who only play for a couple of hours each day at most and don't want to spend that time searching for people to help them complete a mission.
Many, many more players are soloists. People who enjoy grouping but for the most part want to adventure on their own. And though MMORPGS are multiplayer games- video games are solo by nature. So while a majority of the playerbase likes the IDEA that other live beings are running around and they can play with their friends they HATE the fact that they always have to be in a group to accomplish anything.
Lastly, I have to disagree with your thoughts on NPCs. In fact I think the exact opposite. Everyone wants a "virtual world" in these MMOs but then only program NPCs that do nothing until players engage them. I say NPCs should always be shaping and changing the world- Especially in a game that has a war going on. Cities should be rising and falling- crimes and saves should be happening- NPCs should have a HUGE relationship with each other and probably out number the playerbase in great numbers.
Create a world that works without players- then let the players come in and play in it.
That is the goal for my game.
my two cents.
#30
And that's exactly why the worlds are not engaging enough to withstand the next MMO that is released, and really the reason why people keep jumping from game to game, even though they don't realize it, mostly. In DAoC and WoW, you get chat bubbles with preselected missions, though in DAoC, it was better by a hair because you got a few "conversational" options that gave you background aside from just assigning you a mission. In SWG and AO, you had mission terminals, which is just using an ATM to withdraw a mission, and there were days that I wanted to stick very sharp things in my eyeballs because it was always the same. Though, I was impressed by AO's dungeon generation system- enough so to have a clone of it ready on standby for my project. But, really, it's all the same, they just hash and rehash and give it a flashy name to make you think it's better.
I fully agree here. Sure, you have to throw more hardware at it, and maybe even get creative to push some of those AI clock cycles through (ie, throw two SLI cards in the server and use them to chomp AI shaders instead of fragment or pixel shaders, return the results in a memory-resident texture that both the shader and the game access and modify, and you have two programmable AI chips), but it's doable. It takes a higher-order AI than anything that's out now, or probably even coming out soon. And it takes research in other disciplines like psychology, literature, and philosophy in order to tweak things so NPC's do the right and wrong things believably. But I think the beginings of that kind of system are completely within our reach, even (especially) as indies, because we can afford to risk more.
If you couple that with the AI above, you'd have to throw serious hardware at it, and it would be worth looking at things like distributed computing to handle the zones.
06/24/2006 (1:08 pm)
Quote:Everyone wants a "virtual world" in these MMOs but then only program NPCs that do nothing until players engage them.
And that's exactly why the worlds are not engaging enough to withstand the next MMO that is released, and really the reason why people keep jumping from game to game, even though they don't realize it, mostly. In DAoC and WoW, you get chat bubbles with preselected missions, though in DAoC, it was better by a hair because you got a few "conversational" options that gave you background aside from just assigning you a mission. In SWG and AO, you had mission terminals, which is just using an ATM to withdraw a mission, and there were days that I wanted to stick very sharp things in my eyeballs because it was always the same. Though, I was impressed by AO's dungeon generation system- enough so to have a clone of it ready on standby for my project. But, really, it's all the same, they just hash and rehash and give it a flashy name to make you think it's better.
Quote:I say NPCs should always be shaping and changing the world- Especially in a game that has a war going on. Cities should be rising and falling- crimes and saves should be happening- NPCs should have a HUGE relationship with each other...
I fully agree here. Sure, you have to throw more hardware at it, and maybe even get creative to push some of those AI clock cycles through (ie, throw two SLI cards in the server and use them to chomp AI shaders instead of fragment or pixel shaders, return the results in a memory-resident texture that both the shader and the game access and modify, and you have two programmable AI chips), but it's doable. It takes a higher-order AI than anything that's out now, or probably even coming out soon. And it takes research in other disciplines like psychology, literature, and philosophy in order to tweak things so NPC's do the right and wrong things believably. But I think the beginings of that kind of system are completely within our reach, even (especially) as indies, because we can afford to risk more.
Quote:...and probably out number the playerbase in great numbers.
If you couple that with the AI above, you'd have to throw serious hardware at it, and it would be worth looking at things like distributed computing to handle the zones.
#31
Anyway, I'll start with some responses.
Kim, it's not so much that I trust the playerbase. To say the 'playerbase' I assume that you mean everybody. This is untrue. Completely the opposite, actually. Who I'm trusting are the hardkore gamers; the ones that have always wanted to be some great hero or leader, and think this idea is cool. Your average player will be content enough to wander dungeons and kill things, and pretty efficiently solo. This is an inescapable component of rpgs, since it's their foundation. Granted. But, for those that take the game seriously, the fruits will be much sweeter. The 'story' missions will be difficult, and will require that the players be unified. If that's not your thing, just don't go to the frontlines. You can still contribute to the other players simply by dungeon crawling. Find lost relics. Let the warriors make war. you're soloing, bitches.
My disdain for npcs is basically that a programmed amalgam of a human will never be as unpredictable or creative as a human. And instead of attempt to mock humanity I try to find out exactly what the significance of an npc is. (Here comes philosophy) an npc is basically a parameter. A source of direction in a chaotic world. They're motivated by rules, such as; generally, "this needs to be done, but like this". What I'm trying to do is extend those rules onto the player, so that they are the parameters. The guidance. Because dynamacy will never come from silicon chips. Just won't happen. If you want dynamic, the human aspect not only has to be emphasized, but exclusive. I equate taking a quest from an npc like reading a book. It's right there. It'll always be the same every time you read it. Even with super-advanced AI, the variables involved are so minute as compared to the emotional and philosophical capacities of the human mind, it just seems like a no-brainer choice to me. A computer will never write a play like Shakespeare could... so I leave that to the sentient beings. The ones capable of lies and creativity. That's what makes a truly dynamic story. What can an npc do that a human player can't? What can a human player do that an npc can't? The choice to me, seems obvious. "Use that which works, discard that which is useless." (The quote goes out to the JKD fans out there! You know who you are.)
Ted, like I mentioned earlier, there is nothing that human can't do that an npc can, and better. Want cities to truly rise and fall? Leave it to the players. Natural selection is not just a biological concept. Essentially, you want to build an mmo like a miniature universe. Design the rules, the perameters. How A interacts with B and C. Then introduce sentience and see what happens. Familiar with the term emergence? Saving a rather large explanation, that's what I'm looking for. Order disguised as complete chaos. The only factor, the only true epitome of importance is motivation. We want to give the players a reason to react realistically, so we set realistic consequences for thier actions. Last time I saw someone die (Dec 15.) he didn't spawn back at medical with a bit of memory loss. His parts were floating around in the ocean and we had to bring him onboard in duffel bags.
But again, there's the pinch. One the one hand, players want interactive fiction. On the other, they want fun. Nobody like to take thier level 75 uberguy and then have someone mobkill them and have to start all the way over again. Fiction is rarely enjoyable for all parties involved. If it was, it would be crappy fiction. The only way to acheive both goals is compromise. Want a simultaneous story? It's going to be hard for newbs. Want character-motivated role playing? You're going to need to have people die. For good. Comprimise, fellas. Somebody's going to get screwed.
Oh, and about the simultaneous story thing... nobody's really talked about it. Any opinions?
06/24/2006 (2:47 pm)
Awesome. Isn't creative debate helpful? Hooray for Socrates!Anyway, I'll start with some responses.
Kim, it's not so much that I trust the playerbase. To say the 'playerbase' I assume that you mean everybody. This is untrue. Completely the opposite, actually. Who I'm trusting are the hardkore gamers; the ones that have always wanted to be some great hero or leader, and think this idea is cool. Your average player will be content enough to wander dungeons and kill things, and pretty efficiently solo. This is an inescapable component of rpgs, since it's their foundation. Granted. But, for those that take the game seriously, the fruits will be much sweeter. The 'story' missions will be difficult, and will require that the players be unified. If that's not your thing, just don't go to the frontlines. You can still contribute to the other players simply by dungeon crawling. Find lost relics. Let the warriors make war. you're soloing, bitches.
My disdain for npcs is basically that a programmed amalgam of a human will never be as unpredictable or creative as a human. And instead of attempt to mock humanity I try to find out exactly what the significance of an npc is. (Here comes philosophy) an npc is basically a parameter. A source of direction in a chaotic world. They're motivated by rules, such as; generally, "this needs to be done, but like this". What I'm trying to do is extend those rules onto the player, so that they are the parameters. The guidance. Because dynamacy will never come from silicon chips. Just won't happen. If you want dynamic, the human aspect not only has to be emphasized, but exclusive. I equate taking a quest from an npc like reading a book. It's right there. It'll always be the same every time you read it. Even with super-advanced AI, the variables involved are so minute as compared to the emotional and philosophical capacities of the human mind, it just seems like a no-brainer choice to me. A computer will never write a play like Shakespeare could... so I leave that to the sentient beings. The ones capable of lies and creativity. That's what makes a truly dynamic story. What can an npc do that a human player can't? What can a human player do that an npc can't? The choice to me, seems obvious. "Use that which works, discard that which is useless." (The quote goes out to the JKD fans out there! You know who you are.)
Ted, like I mentioned earlier, there is nothing that human can't do that an npc can, and better. Want cities to truly rise and fall? Leave it to the players. Natural selection is not just a biological concept. Essentially, you want to build an mmo like a miniature universe. Design the rules, the perameters. How A interacts with B and C. Then introduce sentience and see what happens. Familiar with the term emergence? Saving a rather large explanation, that's what I'm looking for. Order disguised as complete chaos. The only factor, the only true epitome of importance is motivation. We want to give the players a reason to react realistically, so we set realistic consequences for thier actions. Last time I saw someone die (Dec 15.) he didn't spawn back at medical with a bit of memory loss. His parts were floating around in the ocean and we had to bring him onboard in duffel bags.
But again, there's the pinch. One the one hand, players want interactive fiction. On the other, they want fun. Nobody like to take thier level 75 uberguy and then have someone mobkill them and have to start all the way over again. Fiction is rarely enjoyable for all parties involved. If it was, it would be crappy fiction. The only way to acheive both goals is compromise. Want a simultaneous story? It's going to be hard for newbs. Want character-motivated role playing? You're going to need to have people die. For good. Comprimise, fellas. Somebody's going to get screwed.
Oh, and about the simultaneous story thing... nobody's really talked about it. Any opinions?
#32
Let's say you're going to do a mmo and the setting is a war between Humans and Parasitical Aliens.
You have all Players in all the roles. You have left it up to them how many Parasitical Aliens exist- how many humans exist. What parts of the map(s) are worthy of fighting over- heck, you've even left the motivation for fighting up to them.
What I'm talking about is a world that- even if there was no one playing- would have armies of NPCs taking over areas- making deals with each other- etc. Much in the same way they do in RTSs like Civ 4.
So in the example listed above- The Humans and the Parasites would be at war- but you would see it! The parasites would make active attempts at taking human cities and vice versa.
The players are then affected because NPC armies are invading their towns and changing their lives. NOW they not only experience a dynamic setting they are also motivated to participate.
Now, as you said- and I agree- PCs are smarter/better than NPCs so they will still determine in a huge way which way the universe will turn. But in what I'm talking about- the NPCs won't let it stay that way for long. So whether you solo- group- powergame- casual play- whatever, the server is always throwing something at you.
As far as generating quests and static quests- I'm not big on either of them but I can see how they are necessary on some level- still trying to work around all that.
My point is, leaving your setting up to gamers...even the hardcore gamers- is a mistake IMO.
06/24/2006 (4:27 pm)
I hear what you are saying about NPCs not comparing to humans- I don't think you can do that either and that's not really what I'm trying to say. And I'm not talking about static quest either.Let's say you're going to do a mmo and the setting is a war between Humans and Parasitical Aliens.
You have all Players in all the roles. You have left it up to them how many Parasitical Aliens exist- how many humans exist. What parts of the map(s) are worthy of fighting over- heck, you've even left the motivation for fighting up to them.
What I'm talking about is a world that- even if there was no one playing- would have armies of NPCs taking over areas- making deals with each other- etc. Much in the same way they do in RTSs like Civ 4.
So in the example listed above- The Humans and the Parasites would be at war- but you would see it! The parasites would make active attempts at taking human cities and vice versa.
The players are then affected because NPC armies are invading their towns and changing their lives. NOW they not only experience a dynamic setting they are also motivated to participate.
Now, as you said- and I agree- PCs are smarter/better than NPCs so they will still determine in a huge way which way the universe will turn. But in what I'm talking about- the NPCs won't let it stay that way for long. So whether you solo- group- powergame- casual play- whatever, the server is always throwing something at you.
As far as generating quests and static quests- I'm not big on either of them but I can see how they are necessary on some level- still trying to work around all that.
My point is, leaving your setting up to gamers...even the hardcore gamers- is a mistake IMO.
#33
I can mention a few that are debatable, but there is one that is hands down the single ability that NPC's have over PC's in MMO's that I don't see us being able to take away from them without a severe risk of our real lives: They can stay logged into the game indefinitely
And there is the mistake that I believe you're making in relying on PC's to take over completely for NPC's. Not that the ideal isn't great- in an ideal MMO, the good and bad guys would all be PC's, with no NPC's existing. But then, PC's have two problems: 1) We have to eat/sleep/talk to the husband or wife, etc., as well as respond to "emergencies" that come up. NPC's don't have to do any of those things, except in the context of the game, which is their "real life"; and 2) Not everyone will be keen on playing along as you want them to. Sure, you can put the game rules in place to steer them in that direction, but what's to say that there won't be PK'ing of friendlies in some areas just to disrupt the game there, or that an exploit in the rules will severely unbalance the game? These are things you need to address comprehensively to avoid problems when your game launches. Or, at the very least have contingencies planned.
Remember, hardcore gamers are not all dedicated to playing within the rules. Some of them, no matter how small the percentage is, will buck the system.
See above. People don't like being used as parameters, and will change the rules of the game by gaming them.
And more to the point- are the Fallen PC's or NPC's? If they're not PC's, then you need to work on the AI that's needed for them to do their part, if you want a true give and take. If they're PC's, then your worries fall to the previous point I brought up- but at least you won't need to code a lot of AI in the game ;)
Depends on what you mean by simultaneous stories. Do you mean multiple missions, or branching outcomes, or something else?
Socrates = teh win!!11!!1! hehe..
06/24/2006 (6:16 pm)
Quote:like I mentioned earlier, there is nothing that human can't do that an npc can, and better.
I can mention a few that are debatable, but there is one that is hands down the single ability that NPC's have over PC's in MMO's that I don't see us being able to take away from them without a severe risk of our real lives: They can stay logged into the game indefinitely
And there is the mistake that I believe you're making in relying on PC's to take over completely for NPC's. Not that the ideal isn't great- in an ideal MMO, the good and bad guys would all be PC's, with no NPC's existing. But then, PC's have two problems: 1) We have to eat/sleep/talk to the husband or wife, etc., as well as respond to "emergencies" that come up. NPC's don't have to do any of those things, except in the context of the game, which is their "real life"; and 2) Not everyone will be keen on playing along as you want them to. Sure, you can put the game rules in place to steer them in that direction, but what's to say that there won't be PK'ing of friendlies in some areas just to disrupt the game there, or that an exploit in the rules will severely unbalance the game? These are things you need to address comprehensively to avoid problems when your game launches. Or, at the very least have contingencies planned.
Remember, hardcore gamers are not all dedicated to playing within the rules. Some of them, no matter how small the percentage is, will buck the system.
Quote:an npc is basically a parameter. A source of direction in a chaotic world. They're motivated by rules, such as; generally, "this needs to be done, but like this". What I'm trying to do is extend those rules onto the player, so that they are the parameters.
See above. People don't like being used as parameters, and will change the rules of the game by gaming them.
And more to the point- are the Fallen PC's or NPC's? If they're not PC's, then you need to work on the AI that's needed for them to do their part, if you want a true give and take. If they're PC's, then your worries fall to the previous point I brought up- but at least you won't need to code a lot of AI in the game ;)
Quote:Oh, and about the simultaneous story thing... nobody's really talked about it. Any opinions?
Depends on what you mean by simultaneous stories. Do you mean multiple missions, or branching outcomes, or something else?
Quote:Awesome. Isn't creative debate helpful? Hooray for Socrates!
Socrates = teh win!!11!!1! hehe..
#34
Imagine you have a medieval-style mmo. In a typical small town, you would find a bunch of villagers going about some tasks. Each villager would have its own AI to control its behavior, according to whatever its job is, be that look busy farming, defend the town by fighting off any mobs, or try to get players to help with a 'quest' that needs doing. These would "mobile" AIs, since they are attached to objects that can move about the world. There could also be group-level AIs that could manage a group of AI soldiers that are fighting off goblin raiders, who each have their own AI as well as a group AI managing them. These group AIs would also be mobile.
All the villagers in the town, however, could be controlled by a Town AI, which would be a form of "fixed" AI since it is attached to a specific location in the world (though maybe if the town was destroyed and the villagers all had to relocate, it could become mobile? In that case "fixed" would simply mean it spends most of its time in one place). The Town AI would manage assigning tasks to individual villager and group AIs. One level up, maybe the town is part of a small region, which could have its own AI, and that region is allied with a certain faction, like a kingdom, which would have its own global AI. A system of relationships between different levels of AIs could manage allegiances, for example if one town starts to have a better relationship with a different faction (say through player interaction) it could switch itself into the AI hierarchy of the other faction.
For the world to really be dynamic, the AIs would need to interact and respond to each other. Here is my example: Say Kingdom X and Kingdom Y are neighbors and dont like each other very much. Kingdom Y has lots of soldiers raiding towns on the border with X. Eventually this triggers a flag in the faction AI for Kingdom X that something needs to be done. It needs soldiers to fight Kingdom Y, but its army isn't large enough, so it needs to conscript villagers from its domain to send to the front. But the faction AI is only had a direct connection to the capital city AI, so it first orders them to send some soldiers to the front, then orders it to send messengers to outlying towns to recruit their soldiers to send to the front. The Capital City then implements those orders using its own AIs. A big city could have a separate AI for each sub-section. Maybe the Marketplace AI responds to the order by creating more quests for gathering resources for the war effort, and shortages of some goods. The military section would spawn groups of soldiers to send to the front. It could also spawn messengers that need an escort to their towns, that players could provide. Or the players could carry sealed orders to the town themselves.
When the messengers reach a town, they convey the Faction orders to the town. If the messenger is killed along the way, the town doesn't get the message, and eventually the faction AI will send a new messenger. The town AI can respond to the order by changing its objective: Now it needs to build a training center. It can spawn an under construction building, and issue orders to villagers to gather wood for construction materials. But this brings the villagers into contact with trolls in the woods, which leads to a higher frequency of villager deaths due to the trolls. The Town AI could respond by issuing quests to players to kill the trolls. This is where quests become more interesting. If the whole chain of events has been stored as part of the AI objectives, it can be included in the quest description. Instead of "Kill 5 Trolls" the quest can become "Kill the trolls to protect our villagers. They need to gather wood to build our training center, so we can send soldiers to fight against Kingdom Y". WAY more interesting. Furthermore, the outcome of the conflict between Kingdom X and Kingdom Y can be determined by what players do to help each faction. This can carry through down to the fate of individual towns on the border. If players help in various areas, maybe the town will be saved. If players help Kingdom Y, maybe it will be burned to the ground.
This isn't a perfect system, because the AI complexity would become quite large and would take a lot of effort to program. But it seems to me a relatively simple way (from a design point of view, not implementation) to introduce more dynamic worlds where players' actions can have consequences and the world will seem more believable, without the need of content writers scripting everything that happens. Again, sorry for the long post.
06/28/2006 (3:49 pm)
In regards to generating dynamic stories and having a world that runs itself, I had an idea a while ago about using multiple levels of AI to simulate a dynamic world. I think I can best illustrate this with an example (I apologize for the long post, I've been thinking about this for a while but haven't documented it yet).Imagine you have a medieval-style mmo. In a typical small town, you would find a bunch of villagers going about some tasks. Each villager would have its own AI to control its behavior, according to whatever its job is, be that look busy farming, defend the town by fighting off any mobs, or try to get players to help with a 'quest' that needs doing. These would "mobile" AIs, since they are attached to objects that can move about the world. There could also be group-level AIs that could manage a group of AI soldiers that are fighting off goblin raiders, who each have their own AI as well as a group AI managing them. These group AIs would also be mobile.
All the villagers in the town, however, could be controlled by a Town AI, which would be a form of "fixed" AI since it is attached to a specific location in the world (though maybe if the town was destroyed and the villagers all had to relocate, it could become mobile? In that case "fixed" would simply mean it spends most of its time in one place). The Town AI would manage assigning tasks to individual villager and group AIs. One level up, maybe the town is part of a small region, which could have its own AI, and that region is allied with a certain faction, like a kingdom, which would have its own global AI. A system of relationships between different levels of AIs could manage allegiances, for example if one town starts to have a better relationship with a different faction (say through player interaction) it could switch itself into the AI hierarchy of the other faction.
For the world to really be dynamic, the AIs would need to interact and respond to each other. Here is my example: Say Kingdom X and Kingdom Y are neighbors and dont like each other very much. Kingdom Y has lots of soldiers raiding towns on the border with X. Eventually this triggers a flag in the faction AI for Kingdom X that something needs to be done. It needs soldiers to fight Kingdom Y, but its army isn't large enough, so it needs to conscript villagers from its domain to send to the front. But the faction AI is only had a direct connection to the capital city AI, so it first orders them to send some soldiers to the front, then orders it to send messengers to outlying towns to recruit their soldiers to send to the front. The Capital City then implements those orders using its own AIs. A big city could have a separate AI for each sub-section. Maybe the Marketplace AI responds to the order by creating more quests for gathering resources for the war effort, and shortages of some goods. The military section would spawn groups of soldiers to send to the front. It could also spawn messengers that need an escort to their towns, that players could provide. Or the players could carry sealed orders to the town themselves.
When the messengers reach a town, they convey the Faction orders to the town. If the messenger is killed along the way, the town doesn't get the message, and eventually the faction AI will send a new messenger. The town AI can respond to the order by changing its objective: Now it needs to build a training center. It can spawn an under construction building, and issue orders to villagers to gather wood for construction materials. But this brings the villagers into contact with trolls in the woods, which leads to a higher frequency of villager deaths due to the trolls. The Town AI could respond by issuing quests to players to kill the trolls. This is where quests become more interesting. If the whole chain of events has been stored as part of the AI objectives, it can be included in the quest description. Instead of "Kill 5 Trolls" the quest can become "Kill the trolls to protect our villagers. They need to gather wood to build our training center, so we can send soldiers to fight against Kingdom Y". WAY more interesting. Furthermore, the outcome of the conflict between Kingdom X and Kingdom Y can be determined by what players do to help each faction. This can carry through down to the fate of individual towns on the border. If players help in various areas, maybe the town will be saved. If players help Kingdom Y, maybe it will be burned to the ground.
This isn't a perfect system, because the AI complexity would become quite large and would take a lot of effort to program. But it seems to me a relatively simple way (from a design point of view, not implementation) to introduce more dynamic worlds where players' actions can have consequences and the world will seem more believable, without the need of content writers scripting everything that happens. Again, sorry for the long post.
#35
Another thing relating to reputation being transferred as villagers travel is "talk of" things, such as war. If villagers in one town have been warned about impending war, and villagers in another have not, then when the villagers from town A go to town B, then those in town B will hear about the warnings and begin to become apprehensive, though they would not start preparing nearly as fast as if they had gotten the message from the kingdom's messenger. It would be a far slower build-up, like a small increase in the night watch, followed by sentries in the daytime, then a material ramp-up, fortifications being built, and then training and patrols. But those stages of preparation would be slower, probably also depending on the distance between towns. After all, two towns five miles apart would react a lot faster to developments in each other than two towns fifty miles apart.
But it's a good way of handling things, and using the AI in that manner allows the NPC's to play it out like an RTS at the higher levels, with the PC's acting as independent units. This goes right to what I was saying earlier, about how the AI and story have to be integrated into every aspect of the game, and not just at the "quest description text" level.
06/29/2006 (6:58 am)
That's definitely one of the better ways to do it: Instead of hiding things like messages being sent to areas, actually spawning and sending a messenger to deliver it, with the chance of getting killed along the way. You can also tie it into a reputation system so that if a player is "seen" fighting Ogres or saves villagers gathering lumber from those Ogres, they gain reputation with that village, and as the villagers travel to other towns, some of that reputation is transferred so that the player's reputation proceeds them when they arrive at one of those other towns.Another thing relating to reputation being transferred as villagers travel is "talk of" things, such as war. If villagers in one town have been warned about impending war, and villagers in another have not, then when the villagers from town A go to town B, then those in town B will hear about the warnings and begin to become apprehensive, though they would not start preparing nearly as fast as if they had gotten the message from the kingdom's messenger. It would be a far slower build-up, like a small increase in the night watch, followed by sentries in the daytime, then a material ramp-up, fortifications being built, and then training and patrols. But those stages of preparation would be slower, probably also depending on the distance between towns. After all, two towns five miles apart would react a lot faster to developments in each other than two towns fifty miles apart.
But it's a good way of handling things, and using the AI in that manner allows the NPC's to play it out like an RTS at the higher levels, with the PC's acting as independent units. This goes right to what I was saying earlier, about how the AI and story have to be integrated into every aspect of the game, and not just at the "quest description text" level.
#36
How about making perma-death as an option? That way, when someone is tired of a character and wants to start another one or when they get tired of the game and want to move on to another one, they can go out in a blaze of glory.
06/29/2006 (7:38 am)
I haven't read the whole thread so maybe someone already mentioned this. People worry that perma-death may drive people away from the game but not having it has the same effect on the long run. People get tired and move on to the next MMO only after a few months. In essence they kill their character when they quit the game.How about making perma-death as an option? That way, when someone is tired of a character and wants to start another one or when they get tired of the game and want to move on to another one, they can go out in a blaze of glory.
#37
06/29/2006 (12:22 pm)
Another thing you could do to improve realism is have the towns keep a history of the major events that happen there, and maybe who was involved in them. Then each player can have a log of when they last visited a town, and when they return the villagers could talk to them about things that have happened since they left. Wouldn't it be great if you helped save some village from an Ogre attack and then moved on, and while you are elsewhere you hear from some player that there was a big battle there and the Ogres were driven off completely, and the next time you return to the town the villagers tell you about it? That way a players actions aren't only evident in the results, like village being saved, but it will even be incorporated into conversations the villagers have with other players.
#38
Point is, you can expand the memory system the same way as the AI system. Of course, the next logical step is the alignment/reputation system, where individuals and groups would have a frame of reference for PC's that cross their path more than once, if they do anything of consequence near enough to them for the NPC to "see" it. From there, a weighted average of the PC's works (good, bad, or otherwise) is compiled and the NPC can adjust it's behavior to the PC based on that.
For example: A Theif comes to a town and steals a piece of bread from a vendor, but gives two pieces of copper to a beggar walking out of the store. Another NPC sees the act and does not interfere, but builds a frame of reference that the theif is unlawful and good. Down the road, the Theif comes across a damsel in distress and rescues her, and that NPC builds a frame of reference that the Theif is good and lawful. Yet further down the road, the NPC's brother is mugged by said theif, and earns the "evil, unlawful" frame of reference (now, for references, you can use whatever you want to convey what you need to). The next time the Theif comes to town, he needs the help of the damsel, who he "knows" he has helped and has a good opinion of him. However, when he approaches her, this NPC's frame of reference has changed for the worse, weighted by the communication of her brother's frame of reference about the Theif's actions. The Theif then goes to the NPC which saw him steal, and that NPC helps him, knowing that he stole, but also knowing that he does good things as well (and in this particular case, the NPC does not exchange frame of reference data with the other two NPC's, so his did not get any worse than what it was to begin with).
These frames of reference are relevant to averaged values over a certain span of time, and are nowhere near as complex as our own memory references. You can, however, probably get something more complex by using multiple "frames", and linking that to an event database, so the NPC knows "why" it thinks the Theif was evil on a certain occasion.
06/29/2006 (1:45 pm)
Not only that, but you can store minor events as well, to help players on quests. If a player is looking for the Dire Curmudgeon of Darkness, and it passes through an area every once in a while, it would be helpful to ask locals if they've seen it. Of course, some locals are old and senile, and may have forgotten that it passed by, where as others would have the passing burned into their memories for a very long time because he stepped on their house and pruned a few branches off of their family tree.Point is, you can expand the memory system the same way as the AI system. Of course, the next logical step is the alignment/reputation system, where individuals and groups would have a frame of reference for PC's that cross their path more than once, if they do anything of consequence near enough to them for the NPC to "see" it. From there, a weighted average of the PC's works (good, bad, or otherwise) is compiled and the NPC can adjust it's behavior to the PC based on that.
For example: A Theif comes to a town and steals a piece of bread from a vendor, but gives two pieces of copper to a beggar walking out of the store. Another NPC sees the act and does not interfere, but builds a frame of reference that the theif is unlawful and good. Down the road, the Theif comes across a damsel in distress and rescues her, and that NPC builds a frame of reference that the Theif is good and lawful. Yet further down the road, the NPC's brother is mugged by said theif, and earns the "evil, unlawful" frame of reference (now, for references, you can use whatever you want to convey what you need to). The next time the Theif comes to town, he needs the help of the damsel, who he "knows" he has helped and has a good opinion of him. However, when he approaches her, this NPC's frame of reference has changed for the worse, weighted by the communication of her brother's frame of reference about the Theif's actions. The Theif then goes to the NPC which saw him steal, and that NPC helps him, knowing that he stole, but also knowing that he does good things as well (and in this particular case, the NPC does not exchange frame of reference data with the other two NPC's, so his did not get any worse than what it was to begin with).
These frames of reference are relevant to averaged values over a certain span of time, and are nowhere near as complex as our own memory references. You can, however, probably get something more complex by using multiple "frames", and linking that to an event database, so the NPC knows "why" it thinks the Theif was evil on a certain occasion.
#39
I want to step back a second pull a partial hijack and talk about using RTS tatics with NPCs in a game where PCs play individuals because that is very much the game I am trying to make.
I'm designing a game where the NPCs live in cities that work like RTSs such as Civ4 or SIMCITY. With armies that move/attack/defend in the same manner.
Players will work their way up to play the game on that level as well (work to become an officer and have control of a squadron of troops- or as a general you could control a base or a fleet. As a mayor you would control what buildings are added to your city- bases, water towers, hospitals-etc. etc.).
The biggest problem I'm having to work out is the pace- RTSs have a lot of development happen in a few minutes (such as the building of an entire nation in a couple of hours) whereas RPGs are a lot closer to real time (which is kind of ironic).
1. Is that possible with Torque? I imagine it is because Torque makes RTSs as well as RPGs- there's gotta be a way to mix the two.
2. Is it feasible? With thousands of players potentially controlling tens of NPCs in a universe of NPCs running around freely- am I asking too much of the servers?
06/30/2006 (7:55 am)
That's interesting and sounds complex as hell, Ted- which is a good thing. I want to step back a second pull a partial hijack and talk about using RTS tatics with NPCs in a game where PCs play individuals because that is very much the game I am trying to make.
I'm designing a game where the NPCs live in cities that work like RTSs such as Civ4 or SIMCITY. With armies that move/attack/defend in the same manner.
Players will work their way up to play the game on that level as well (work to become an officer and have control of a squadron of troops- or as a general you could control a base or a fleet. As a mayor you would control what buildings are added to your city- bases, water towers, hospitals-etc. etc.).
The biggest problem I'm having to work out is the pace- RTSs have a lot of development happen in a few minutes (such as the building of an entire nation in a couple of hours) whereas RPGs are a lot closer to real time (which is kind of ironic).
1. Is that possible with Torque? I imagine it is because Torque makes RTSs as well as RPGs- there's gotta be a way to mix the two.
2. Is it feasible? With thousands of players potentially controlling tens of NPCs in a universe of NPCs running around freely- am I asking too much of the servers?
#40
As for the RTS aspects in an RPG world, it's something that I've been thinking on for years, and the time compression is not so much a problem, as long as the NPC side of the gameworld behaves a lot like the PC side. And example is that in my game, I want there to be community and organizational (guild) projects, such as building a base or repairing the walls of a city. This is where PC's get to use tradeskills to do something for others as well as themselves. In the gameworld, a group of motivated players will be able to build a base in a shorter period of time than the real world, but a longer period of time than in an RTS, which usually only takes a minute or two. What you could do is average out the build times as compared to how PC's would build their bases, and translate that to the NPC's, so that you have some sort of level playing field on that front.
The AI for the moving of troops and whatnot? Well, that's a bit more complicated, and here's why: PC's don't play by the rules, and you're going to be sending NPC's into areas where they need to confront and destroy the enemy, but also follow along with the plan. So, you need to know when your NPC's are going "off plan" and put in controls to reign them in. Sometimes, you will want them to pursue characters who are running away, and sometimes you will not want them to.
Maybe a good system would be a hierarchical AI that was discussed, but instead of an all-controlling AI, you would have a division between strategic and tactical AI. Strategic AI would be the higher portions of the hierarchy, doling out high-level tasks to the lower-level AI to accomplish, but not managing their affairs (or not micromanaging them), while the tactical AI would be the "unit commanders" that represent groups of lesser AI "components". You can, in effect, use this to design a model wherein the AI unit commander is subject to many of the limitations that the PC's would have, and therefore subject to the tactics that can defeat him in a similar way (while also having the same tricks up his/her sleeve, as well as a limited ability to recognize that fact).
By subjecting them to these rules and different levels of AI, you can get more realistic behavior with less overhead. If you implement an "LOD AI" approach, then you can allow for the AI to ramp up when it needs to on a particular NPC, and then almost completely shut off if there is noone around, and use quick and dirty calculations to decide outcomes of actions. Also, using LOD AI, you can have a relatively minor NPC with basic AI rise up the ranks as unit commanders above it get killed, taking on more responsibility and issuing orders to it's subordinates much like is done in real life.
06/30/2006 (8:42 am)
The answer to both questions is simply: What are you capable of? Of course, there are things that TGE/TSE cannot do in stock, and you'll need to do some coding and/or scripting to accomplish that, but that's true for most engines out there. The RTS pack is mostly geared towards trimming down the "weight" of the Player character class for better network performance, as well as some GUI enhancements. Lessons can be learned from what was done in that pack to help you trim down the network usage. You also have to think creatively on how to get things done with less clock cycles or bandwidth. It's all cheating and slight of hand that makes it look real- every bit of AI does not necessarily need to be processed in the world. You'll probably want to look up concepts like Level Of Detail AI to help with your cause. Whether you're asking too much of your servers just depends on what you're specifically asking of them.As for the RTS aspects in an RPG world, it's something that I've been thinking on for years, and the time compression is not so much a problem, as long as the NPC side of the gameworld behaves a lot like the PC side. And example is that in my game, I want there to be community and organizational (guild) projects, such as building a base or repairing the walls of a city. This is where PC's get to use tradeskills to do something for others as well as themselves. In the gameworld, a group of motivated players will be able to build a base in a shorter period of time than the real world, but a longer period of time than in an RTS, which usually only takes a minute or two. What you could do is average out the build times as compared to how PC's would build their bases, and translate that to the NPC's, so that you have some sort of level playing field on that front.
The AI for the moving of troops and whatnot? Well, that's a bit more complicated, and here's why: PC's don't play by the rules, and you're going to be sending NPC's into areas where they need to confront and destroy the enemy, but also follow along with the plan. So, you need to know when your NPC's are going "off plan" and put in controls to reign them in. Sometimes, you will want them to pursue characters who are running away, and sometimes you will not want them to.
Maybe a good system would be a hierarchical AI that was discussed, but instead of an all-controlling AI, you would have a division between strategic and tactical AI. Strategic AI would be the higher portions of the hierarchy, doling out high-level tasks to the lower-level AI to accomplish, but not managing their affairs (or not micromanaging them), while the tactical AI would be the "unit commanders" that represent groups of lesser AI "components". You can, in effect, use this to design a model wherein the AI unit commander is subject to many of the limitations that the PC's would have, and therefore subject to the tactics that can defeat him in a similar way (while also having the same tricks up his/her sleeve, as well as a limited ability to recognize that fact).
By subjecting them to these rules and different levels of AI, you can get more realistic behavior with less overhead. If you implement an "LOD AI" approach, then you can allow for the AI to ramp up when it needs to on a particular NPC, and then almost completely shut off if there is noone around, and use quick and dirty calculations to decide outcomes of actions. Also, using LOD AI, you can have a relatively minor NPC with basic AI rise up the ranks as unit commanders above it get killed, taking on more responsibility and issuing orders to it's subordinates much like is done in real life.
Nokill
becouse in PvP situations you will get killed pretty often and also if your in a war its not nice if you get a PD all the time so it should be somewhat hard to get killed
and for the leveling its indeed better to leave the lvl status away and try to make evryone equal but when you go mining for instance you will get stronger get better sight of how to mine that mineral etc etc
not that its woesh your a lvl up but that it goe's steady and unnoticed untill you would all up a sudden get back to your first status and the one your at after weeks of training so its not that you can realy feel the difrence like that