Sign in to follow this  

How to make a Dynamic MMO world

This topic is 3726 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

One of my major philosophies for MMO’s is that the player (individually) cannot change the world. This is one reason why I do not believe in letting the players play kings, CEO’s, mob boss’s etc in games. My belief is that players should be able to influence the world through actions of large numbers of players, with reactions coming from the game. I.E. Population control of an area’s mobs. A player should not be able to wipe out an entire population of gnolls, but 80-100 players focused on the task could possibly do it. Please keep that philosophy in mind as I describe my thesis below. An MMO world, should be in its nature interconnected and complex. Players should be able to influence world events as a playerbase, whether they mean to or not. The world around them should dynamically change based on various factors that are influenced by actions the players take in the world, but not tied directly to any one action. Almost every action the players take should be tallied, and stored until like water in an unbalanced bucket, turns over and spills out the consequences. When it comes to world building, a lot of games focus around making 2-3 competing factions, while this is workable, it does not create a dynamic enough playing field. Instead of a coldwar, 2-superpower conflict, I believe the game needs to be more of a middle ages Europe, with multiple nation states, on the verge of competition. This creates a suitable environment for interesting developments. My main method of dynamics is AI driven conflict between nations. The AI does not have to be sufficiently advanced, but each country should have its own “personality.” Conflict should encompass three types of conflict: Trade, Political, and Armed. In order to influence one of the three, players take up quests and missions, as well as just grind monsters, buy and sell goods etc. One must is a dynamic mission generation system. I use the word system because it needs to have multiple front ends, and maybe one or two middle processes, and multiple end processes. The front ends control what the type of mission is, who its for, who its against what you are doing etc. The middle processes take the input from the front ends, and generates the meat of the mission (map [if dungeon], area, enemies, waypoints etc) and the back ends take the results of the mission (success, failure, partial success, side objectives etc) tallies the numbers, and adds them to the conflict counters between nations. For example one front end may deal with political input from competing factions. In the back ground a pool of points is competed over by the factions, and each mission generated and played by the players creates a change in points. This causes more missions to be generated by other factions in order to make up for the loss of points by the first faction. The player chooses a few things like difficulty, length, style and goes off on the mission. Success or failure redistributes points against the pool, resulting in more conflict between these factions. Once a conflict breaks out between different factions, whether it be trade, political, armed etc. A nation should be able to build an alliance with another country to last for the duration of the conflict. Such an alliance would shift numbers and totals needed for a conflict to break out between allied factions making it harder but not impossible for conflict to break out. This also shifts numbers between the allied and warring factions making it much easer for conflict to break between them. Just like the numbers needed to generate a conflict there are numbers needed to end a conflict. These can be political, trade, or warfare as well. Diplomatic overtures may start springing up after a length of time of war, economies could decline, one side loses etc. However, a lot of these numbers, and missions needed to bring these events to an end are going to be generated by players actions. Politics Politics can and should be local and global. The processes should pretty much mirror themselves. The “goal” of the lower level politics is to get an NPC into the leader slot, where their personality (their trade/war/politics modifiers) will affect the nation. I’ll let individual game developers figure out exactly what missions would be needed. Armed conflict When war breaks out, two or more countries start vying for land between them, with battles held periodically. The AI’s determine where to battle (1 or more zones) and players begin completing missions. These missions in turn give each side bonus’s and negatives when the battle takes place. The battle is a zone wide long duration (hours) PVP battle. When the battle is over the winning faction takes (or retains) control of the zone, roving patrols switches factions, prices change as merchants get different suppliers etc. Trade Merchants sell goods. Goods require resources. Resources have to be transported, etc. Players can Gather resources to sell to merchants, or they can rob NPC’s transporting them. The more resources a merchant has, the more goods he can make and the lower the prices. The less…well you get the idea. Also certain goods require certain resources, if the resources are doled out in such a way that not every country has all the resources they need, then trade becomes influenced by, and influences the other two spheres of influence. All of the AI that runs the dynamic world is or should be as transparent to the player as possible. Players should not have to worry about how to get countries to go to war. They will, because that’s what players do, but, individual players should be forced to react to the changing dynamics of the world, not plan to make them happen. Well this is my idea, what do you think?

Share this post


Link to post
Share on other sites
The ability to impact a persistent world will eventually become the next step for MMOs (once they all stop bumping into each other in the copycat frenzy).

I am 100% in favor of impact. A simple example of impact in numbers is the ability to wear a path in the ground after of thousands of people walk over it. This would create trails that represent the quickest path between destinations, and it's all player-created!

If you could somehow make the players impact the world more, you would not have to create as much content, because the players themselves are making some of it!

The first person to implement this, or the best person to implement this, is going to dominate the market next. WoW has too much content to just redesign it's engines for a consumable world, and when someone makes this change to the MMO, I'm guessing that no current MMOs could keep up. Someone please do it!

We discussed procedural quest generation a while back, take a look if you want: http://www.gamedev.net/community/forums/topic.asp?topic_id=458424.

-Humble Hobos

Share this post


Link to post
Share on other sites
H.H.
I read through your post, and I believe we have a major difference of opinion on dynamic worlds.

In yours you want the world to make physical adjustments to the players actions (i.e paths created, trees felled, creatures depopulated etc.) I understand what you are saying and where you are coming from, but my question is what is the value added? Where is the benefit for the game and the player? I put forth that "Unique" experiences are not necessarily the best way for a game to be. Many players play Single player RPGS and have the same experience as everyone else who plays. Uniqueness does not make a game great.

My proposal is is more about changing the state of affairs in the game to create a dynamic atmosphere. The physical world does not change, but the events that the players are attempting to deal with do. If you can instill in your players a sense of atmosphere that they lose themselves in, you will get a much larger bang for your buck when war breaks out between thier country and another. A dynamic world of the type I describe has benefits for every one who plays. The most disadvantaged players may just be the casuals, who if the game changes too much too quickly will end up lost in an everchanging world. THat is a balance issue that needs to make sure that a dynamic world changes at the right pace.

I do agree that if you kill off an NPC that NPC is dead, however I also believe that the game should generate a new NPC to fill his role, maybe not with the same personality or attitude, but to fill the same role. (Shop keep, nobleman, bum etc)

There are two keys in the concept for a dynamic world. First is that the world does not exist for the AVATARS (not players, avatars) but that the AVATARS exist IN the world. Second is that INDIVIDUAL players do not directly influence the world just as a raindrop does not cause a flood.

I do agree that proceduraly generated quests are a must, but I believe that the quests should be generated based on world basis, and NPC goals that are palpable to the player.

Example in the war situation the computer starts generating quests that are raids against enemy camps, supply missions, intelligence gathering etc. The outcome of these missions put forth bonus's that will be applied to the players partaking in the PVP battle to take place down the road. Of course the other side is doing the same thing which cancels out bonus's etc. So the pre-planning (missions generated) have an effect on the game world.

Another example is with politics (Local) A player gets hired by a noble (random mission) to do some sort of task (gift delivery, kill quest, spying etc) the success of which raises the influence of that nobleman. If the nobleman gets enough influence he becomes the new Govenor/king/mayor etc. and his personality now begins to influence the direction and missions that are offered for the Global politics of that nation.

All of these changes come about because of actions that players do as a whole, not individually. Also if all the players left the game the game world would eventually settle back down to a state very similar, though not identical, to the launch as the ripples caused by the comunity smooth out.

Share this post


Link to post
Share on other sites
I love this idea. I have a small contribution to make as well.
Quote:

If the nobleman gets enough influence he becomes the new Govenor/king/mayor etc. and his personality now begins to influence the direction and missions that are offered for the Global politics of that nation.


I think this should be approached differently. I believe that his influence should get him nominated, but then have elections in which players vote for the nobleman they think has the right personality or strategy for the nation. This, of course, is given that the nation in question is democratic. Then, depending on political success, the AI can determine when an early election is called. During wartimes, if certain actions come to pass, the General might take over the nation temporarily, etc.

Just thought I'd add to your idea. And since votes are the best example of many people changing the world together, it fits you idea.

Share this post


Link to post
Share on other sites
Quote:
Original post by AdamGL
I love this idea. I have a small contribution to make as well.
Quote:

If the nobleman gets enough influence he becomes the new Govenor/king/mayor etc. and his personality now begins to influence the direction and missions that are offered for the Global politics of that nation.


I think this should be approached differently. I believe that his influence should get him nominated, but then have elections in which players vote for the nobleman they think has the right personality or strategy for the nation. This, of course, is given that the nation in question is democratic. Then, depending on political success, the AI can determine when an early election is called. During wartimes, if certain actions come to pass, the General might take over the nation temporarily, etc.

Just thought I'd add to your idea. And since votes are the best example of many people changing the world together, it fits you idea.


Not a bad idea, though it would basically be determined by the genre and feel of the overall game. In the game world that I picture this wouldn't work because I'm picturing more of a fantasy-esque world with alot of nobles and kings and such. However if applied to a sci-fi, Roman, modern day world I think it could work.

The question I have is how do you get players to actually care to vote for an AI driven cantidate? Especially if its an AI driven election (i.e. happens anytime the balance tips right) I can see maybe a planned event, but not really on a random basis.

Share this post


Link to post
Share on other sites
Each candidate would have strategy that they advertise. These would be procedurally generated of course, depending on the circumstance. They would grant players bonuses and promise changes to the nation, like tax reductions off of goods (trading gets affected) or to increase the war effort (more missions) or to end the war and such. Depending on the events, they would promise different outcomes, and grant different bonuses to players. If the candidate wins, their promised outcome has more of a chance of happening. Players would vote because the decision would affect their every day level grinding or questing.

And votes wouldn't be too regular, unless of course, total chaos ensues. They should be rare. And these elections would be announced weeks ahead, and advertising would be done, etc. Shopkeepers ask people's opinions on the upcoming election etc. It should be well known, so everyone can take time off work to go vote in game.

And to add one more thing. A group of players could start a mutiny, in which they get NPCs that they would like to be in certain positions of power, and then they go ahead and attack the government. This would take many players to do, so it wouldn't be frequent, only when everyone disagrees and no election is around (or there is no election at all, due to an autocracy). If the players won, then the NPCs the players chose would replace the current NPCs.

Share this post


Link to post
Share on other sites
As far as I can tell, this is NOT a truly dynamic world. This is "merely" a player and event driven, world. In order to get “dynamic”, you have to get both interactivity between players, and between players and the world

I have had another idea, to get a dynamic world. It's quite simple in name, but rather difficult technically.

Replace most AI by HI. Like Human Intelligence.

Most games designers are aware there are games outside of their fields, and are rather inspired by those. They are rightfully so, because the players usually play games other than those the designers designed.

And one field I would like to attract your attention to is browser-based games.

You may not be fully aware of it, if you haven't been researching them intensely, or merely been interested in them, but there are actually a throng of them, in very different domains.

In some games (ie: [url]www.myefarm.com[/url]), you get to play as a farm-owner. You're not an adventurer, but have to make your farm growing and prosperous. You buy patches of land, you seed them, sow them, you graze cattle, you keep hens, you buy stuff and sell your production, knowing fully that you cannot do all the sides of production, unless you invest way more time than a simple browser-based game should require.

In other games, you may have to raise monsters or random animals, breed them, feed them, train them, pitch them against eachothers...

You may even find flash-based games in which you can perform random tailoring games, or surgery games (a la DS or WII doctors...), or whack-a-mole-like hammering games.

Where am I trying to get you to? To a very simple and straightforward solution.

USE HUMAN WORKFORCE AND BRAINTIME TO FUEL YOUR MMO!!!

Tie up the different browser-based databases, and call them your main MMO database.
Define your overall Map first.
Then, put on that map the "Monsters". Players driven, with goals defined by the games ruleset, evolution paths differing from creature to creature (or creature group, more probably...) and what you get is a very large group of different creatures populating your game world, and moving about in order to fulfil PLAYERS' wills.
Then, put on the same map human farmers, and tell them to colonize your game world. You'll have an economy made of people trying to get to the best patches of land, or going to unknown places, and modifying on a daily basis the landscape of your gameworld. Things are already moving.

Now add to your ruleset that there is a possibility of further evolution for your monsters, that they may not remain forever as rats or spiders, but may become flesh-eating monsters, provided they start to feed on other monsters. By doing so, they may earn points, which they will trade for genetic evolutions or for a change of race. Tell them they may feed on cattle too. Humanly grazed cattle. What you're getting to now is INTERACTION, which is the key to any trulyDynamic world.

Now your farmers have new needs. A need for protection from those intrusive and negatively affecting monsters. They need adventurers. So, your adventurers come last into the process. You get the MMO players to get "quests" from real people, negotiating the rewards and conditions potentially in real time, or offer a simple tool to create "quests" for the farmers, or both. Those "quests", at first, will be very simple, and could be defined as your generic "get me twenty seven pelts of red foxes from my field, and I'll give you 54 gold." Those have been seen times and again in EVERRY MMO out. What you MAY add is "grab as many red foxes during the night. But I'll need them alive, and thrown into my neighboor's field. For each fox thrown there, I'll reward you 3 gold..." Now the difficulty has been raised a notch. But the interest of such quest has been raised tenfold!! You're not only affecting the two players interacting together, you're also affecting an unsuspecting third party, radically modifying the landscape of your MMO world local economy.

You may also find a "fetch me that" quest, through "go in the forest and find me a bee swarm" quest. Bee swarms may not be player driven, but the bears living around it would, and each would make MANY points by driving any player away from his home, and protecting his own source of food. So what you have is a situation in which two players conflict, although not on the same level.

You basic adventurer will also have basic needs for equipment, which will have to be player produced (through that famous flash-based game) and rather unique in terms of qualities. A string of hexadecimal letters defining what an object looks like and is rather than where it can be found in the database may do the trick better than what currently exists. Or maybe not... And those "crafters" would have to be tied to pre-existing shops in your Game World. All those would compete for fame, efficiency and customers, but also for shop spots, since the more you're exposed to customers, the more likely you are to get orders... A

Add in the mix the absence of gold for adventurers loot, and limit the gold in existence to whatever amount has been BOUGHT by the different crafters, plus the fact that the crafters cannot move around the world to gather crafting materials and have to rely on adventurers willing to make a buck or two, and you've already got a thriving aconomy. Adventurers will automatically fall in and create servicing companies, moving goods around, and sharing the money. No money comes in or out, it merely goes arounbd, as everybody HAS to get some money. Item decay may be nice too... The ability of crafters and farmers to buy ingame currency with REAL WORLD money would not overthrow the economy at all. It would merely momentarily imbalance the crafters ladder.



As for the ability to individually momentarily imbalance the world, or at least modify its momentum, could be done through the use of contract quests, just like those linking farmers and adventurers. What you want is for someone to go into such place and get you something? issue an order, and people will probably take the command, and deliver. And the beauty of it is that you can probably use the same system for NPCs too, and induce events without the PCs being really aware that they are already taking part in it, or a string of linked quests triggering some other effect.

If you can add different outcomes for your actions, and various states of success, then you've got a winner...

As a side note, I would like to point you towards winning games, corresponding to dynamic worlds, like [url]www.renaissancekingdoms.com[/url], which is a strategy game, a farming game, a social game, a browser-based game, and a trade game, all rolled successfully into one…
You may also visit [url]www.naturalchimie.com[/url], which is a very addictive little flash-based game, in which you get to play a sort of “Doctor-Mario-meets-Collapse” with alchemical components, in order to create gold nuggets. The process is easy to grasp, and could be used in ANY MMO to create potions or buffs, and if the alchemical components were replaced by crafting materials, to create anything at all.

YoHoHo! Puzzle Pirates ([url]www.puzzlepirates.com[/url]) is also a very good example of how flash-based games can be used to create a synergy between people teaming up, and how crafting can be done through playing, rather than macroing…

Share this post


Link to post
Share on other sites
One way to make a dynamic game world is to have multiple levels of game paly. These would in effect be several different games that have access to an underlying database.

For instance, you might have the typical MMO fantasy dungeon romp. The mobs that get killed can influence the incidence of events in a strategy game (the typical "creeps" found in games like Warcraft III).

Now in that RTS game, the player has to build a base and raise troops, etc and this can then influence the economy of the MMO.

There might be no more interaction than described here. The MMO does not see the buildings of the RTS player, and neither does the RTS player see the players of the MMO running around.

They do however, see the effects of their actions as translated through the database.

You might then also add another layer to it and have something like a farm simulation and then this effects the resource production rates that the RTS player has.

The players from the different games also don;t have to have the same teams as the players in the other games (but they can if you want). For instance the farm simulation game might influence the whole world (averaged out over all the farm sim players).

Share this post


Link to post
Share on other sites
Imagine for a while that you play a browser-based monster breeding game. Your monsters have different "needs".

Their first need is to find "shelter". By finding a "shelter", they will define the area in which they will live and perform every other action. That area will be tied to the map database. BY finding a sure shelter, they will earn points to a player, on a regular, per day basis

They need to "feed", that is to find spots in which they can perform the action of finding food. For doing so, they will earn you, the player, points.
Each item of food found may earn points.

They need to "survive" individually first. That is that, for grass-eating animals and insect-eating animals, to avoid being eaten by other animals or killed by humans. By doing so, they will earn points at a regular rate, on a per hour basis. For flesh-eating monsters, that means inflicting damage to anything attacking him or coming into its territory. Avoiding death as long as possible will help gaining points, of course...

But they will also need to "survive" as a species, which is "mate", "thrive" or "reproduce". For any animal, reproducing means earning points to the player. But in order to do so, you have to find a potential mate.

Mates are likely found in places that cover the most basic needs.



What of the points, you might say. What are they for?

They are to evolve through species ladder.

You will start off as a critter. Rats in cities, spiders in forests and caves.
You may then evolve to small rodents. Rabbits, squirrels, the likes.
Then proceed on to grass-eaters. Deers and such.
On to rodent-eaters. Foxes and Badgers.
Then Flesh eaters. Wolves. Lions.And potentially hyenas too.
Then omni-eaters. Bears and Wolverines.

Further along the line of evolution, you may choose to become a goblin, an orc, or a troll, living in groups.

And ultimately, you may become a dragon.

Of course, becoming a dragon would be the ultimate goal of evolutionners, since there is no other step beyond. But in practice, you may not find many people willing to become one. Because...



Points are always invested FOREVER. Investing points in evolution for the specie you're actually playing with may or may not prove a good move, because those points may set you back a little on the ladder, and may not yield much in return. And changing for another species ALWAYS costs many points. When your last member of your group dies, you go back down the ladder to the higher specie you can buy with your remaining points. And although most species ensure that you can almost always find a mate to reproduce with, some don't.

Dragons are lonely and can only attack everything that comes anywhere near. They don't breed. They may evolve as well. And they get to keep all the equipment of the adventurers that came to kill him and failed. The bigger the "treasure", the more adventurers may want to come and try... Which means that, in the end, dragons ALWAYS loose. But setting a record ALWAYS attract players...

With differential ladders, you'll always get people in the correct categories...

Share this post


Link to post
Share on other sites
Quote:
Original post by Merluche
As far as I can tell, this is NOT a truly dynamic world. This is "merely" a player and event driven, world. In order to get “dynamic”, you have to get both interactivity between players, and between players and the world



My question to you is what sort of world do you picture as dynamic? You call mine player and event driven.

In my opinion a dynamic world changes based on player actions, and then the players have to react to a world that's changed. The dynamic in my world is about the political/trade state of the world.

I am not so much for a physically dynamic world as in my opinion a physically dynamic world would eventually be ruined if it wasn't managed properly and a ruined world IS NOT good for business.

Here are my problems with your description of a dynamic world.

1. Atmosphere. What your describing basically puts random creatures on the world. Walking around an area where a lion and a dragon and a kangaroo are all residing with no sense of reason breaks immersion, and therefore drives some players away.

2. Balance. You have player classes, Farmers, Monster makers.....you will have a nightmare figuring out how to balance it out where even a fraction of your playberase is happy. Also how do you balance a farmer with a monster?

3. players as resources. In your game the players you have are another resource to build your world. The more you divide up your resource, farmers, monsters, players, the smaller your pool will be for each which could make the world seem awfully empty if you can't get enough players to fill the void.

Pause for a moment and picture in your head what the playing experience would be like in your game, do you think it would be fun, what would a player experiance be like etc?

Share this post


Link to post
Share on other sites
I think you may have misunderstood my proposition. I wasn't proposing to have a single game where everybody would CHOOSE what he would do. In fact, it's quite the opposite.

I was proposing to have a wide range of games, for smaller markets to enjoy. I mean, www.myefarm.com boasts a full 700,000 players, www.renaissancekingdoms.com is way over 1,500,000 with all three language servers... And I'm not counting the different monster breeder games around... And those need not be the same players as adventurers. Those browswer-based games in fact, do NOT catter to the same playerbase.

And why do I call your game-idea a "player and event driven game"? Because from what I understood of your Original Post, the state of the world would EVOLVE from events, and events are NOT player-created, but created by the devs, and achieved by the players. What I envisioned as a "dynamic" world would be a world which could evolve from the players alone.

You might argue that players left alone might destroy completely a universe. That is true only to some extent. ADVENTURERS left alone in a universe that might react to their chain killing of everything in sight may very well throw the balance away. But when the other parts of the universe start to adapt to their killing by modifying their comportment, BECAUSE THE REST OF THE UNIVERSE IS MANAGED BY PLAYERS WHO ALSO WANT TO WIN, even though there isn't much to win anyway, except bragging rights... In this respect, the universe could be dubbed "dynamic", because the actions of one or many can alter the behavior and game experience of all others, NOT the actions and decisions of devs. (I'll grant you that devs have even more power to alter the game state in this context, but that they may only alter the game to launch it in new directions.

And as far as athmosphere goes, I REALLY wonder how a kangaroo, a lion and a dragon could possibly find themselves in the same vicinity. For a start, kangaroos are grass-eaters living in open range country. Lions are flesh-eaters living in open range country. Dragons are omni-eaters, mainly dwelling on humans, living in mountainous country, preferably not too far from cities and commercial routes. The reasons why kangaroos and lions may find themselves on the same country can be multiple, provided you offer backstory for the fact. But dragons dwelling in African Plains would, in fact, be counter-productive for the player incarnating the dragon. He would not be able to achieve his primary objectives by going there, since his main objectives, in term of points earned and basic needs would be:
1) to feed, that is to find humans to feed upon, preferably some with much gold and/or items. The total value of their loot is equal to the number of points they earn, and which will help them restart for when they are killed. Not going where points are freely distributed is counter-intuitive for the player.
2) to find shelter. And finding a cave in which a dragon can sleep and rest in open range land is pretty difficult. Therefore, the dragon would slowly see its health decay, and the player would try to take it to shelter, away from that land. Moreover, if he has no cave to stash his loot, he won't earn anything AT ALL.
3) To survive as an individual. Meaning he needs to fight off enemies trying to kill him. If he can't get his back to a solid mountain wall, he will be quickly surrounded and killed. Since evolutions would be bought with points, flight would only be bought after earning points, and that means surviving and achieving basic objectives of stashing loot and feeding.

I'll willingly grant you that lions do dwell in mountains. But those are pretty different from african plains lions. They have evolved differently. And evolution only happens in this context through player individual choice.

Hence "dynamic" world...

Share this post


Link to post
Share on other sites

This topic is 3726 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this