To a certain degree this sort of thing has been done in EV Nova. Although the game state does not persist for each play, there are 6 main objectives and it is only possible to complete one in a game. These objectives aren't necessarily mutually exclusive, it would be possible for some of the end points to happen within the same game world. Along with these 6 main objectives there are a large number of smaller missions that can be completed.

As well as the different objectives, in some cases your actions actually change the outcome of the story, although understandably this didn't happen very often.

The story in EV Nova is all text and missions. The missions are written in a scripting type language and only deal with ships, hostility, upgrades and a few other things. This makes it quite easy to actually be able to write a complex and detailed multi-branching storyline. I think a multi-story persistent game would also have to use similar methods to generate content.

It would probably have to be an rpg of sorts, although it might work well in a gta style (continuous world) game (which probably is an rpg of sorts anyway).

I really like the EV series and the style isn't that far from what you are describing. The changes that you can make to the world are:

Killing important people (assasinations)
Helping people / races and
Destroying races / starting wars (governments)
Finding out information about mysterious things (information)

But for the most part, EV Nova only dealt with government happenings, which are the easiest to make content for. That isn't to say that allocation of territory and other stuff has to be boring, government missions in EV Nova were original, involved named people and had you travelling all over the galaxy.

The problem with RPGs is the people seem to assume that the goal must be to save the world but there are many other possible stories that could be pursued. I'm not saying persistant worlds would be ideal for RPGs but they could be used consider for instance that the average rpg player would probably just as happy if not more so to play 4 12 hour plays where their actions in the previous plays impacted the current one, rather then a usual 48 hour single play RPG.


But one avenue I could defentily see for a persistant game world over multiple plays would be in non empire level 4x games. For instance imagine this your play a space 4x games in which you choose a character archtype and the play a 4 years career segement of that type of character. So at the begining of the game earth has just developed the technology to travel to other planets and you can choose one two archtypes explorer or colony leader. As a colony leader your given a colony ship and given 4 years to to establish a colony on another world. So you fly off to another solar system find a sutiable planet and start a colony. After four years of work the colony is not only surving but thriving, you've developed new minning technology and cargo storage methods that have allowed you to begin exporting rare minerals that the planet is rich in.

So now that your 4 years are over you can pick an archtype to start a new play session as or you could contiune a previous character for another 4 years. But since the world is persistant each of your plays continues to build up the universe. Each new colony you create is an additonal colony you can interact with in a new play. Each new piece of technology you discover is some new to use when you start a new play which can open new areas that where prviously unreachable. One certain requirements are met new archtypes would become available. When there are 5 colonies you could choose to be a tycoon who starts with a set some of money and has 4 years to make as much money as you can. The explorer could travel to new worlds searching alien ruins or meeting new civilizations. On and on things go each play opening up new possiblities and challenges to explore.

We've been debating how much to do this with Xenallure. Should we go to the extreme of allowing a war to start in one gameplay, and have it still be going on in the next gameplay? Should we do something more subtle - the NPCs don't remember the PC, but they are biased to like or dislike the PC according to what kind of relationship they had at the end of the previous game? One of the things you can do in the game is upgrade a town adding buildings and such to it - this should probably be persistant, because it would be annoying to have to re-do it each time, right? The question is what will be to most fun and least cheesy for the player? How canwe space the changes out so that they are not all between the first and second gameplay, but what the player can do in the world continues to change enough to be interesting with every new gameplay (or at least the first 4)?

I LOVE the idea of persistent, ongoing worlds and the whole "New Game+" concept (forget which game originally had that, read it on the boards here).

One problem I've always had, though, is that portion of the RPG which is sheer discovery of the new. If I replay after having a fairly advanced character, I begin to know where everything is.

Maybe this isn't so much of a problem, but from replaying Morrowind you can sort of get a taste of this.

If the game world expands, you're going to have a memory issue. If the gameworld keeps changing, you're going to have a playtesting, QA and repetition issue.

But, of course, that's where creativity comes in. [wink]

Quote:

Original post by Wavinator I LOVE the idea of persistent, ongoing worlds and the whole "New Game+" concept (forget which game originally had that, read it on the boards here).

That would be ChronoTrigger, although Vagrant Story and Eternal Eyes among other games also have this feature.

Actually, New Game+ is just the opposite of what's being discussed here. It was the chance to play the exact same game with the old enhanced characters. The discussion seems to be more about starting a new character in a world that's been affected by previous play-throughs.

Wavinator, are you saying you don't like playing a game knowing where everything is (see the "accelerated start" from Fallout 2), or you don't like staying in a world you've already seen the last time you played? I think you could avoid both scenarios by letting some event occur differently and alter the landscape. For instance, if you save Shady Sands and let the Neproplois fall to the mutants, then the next playthrough--maybe three years later--would feature a bigger, better shady sands and a new faction in the Necropolis (or just ruins, maybe). Like you said, this becomes a design obstacle, but if you have a good modular faction system you could just have a few variables (Necropolis=gang-run & barbaric, Shady Sands=Tandi-led & agrarian, BoS=Rhombus-led & high tech, Cathedral=Glow).

It would be a lot of work, no doubt. It would certainly have to be a lot more than making the game for once through and then hoping to tweak it from then on. You'd really have to make a game's worth of material for each play-through. With that said, you might make a play-through only take four hours, so it would be one game's worth of effort yielding six plays' worth of content.

Iron chef Caranage, Wavinator is not wrong. New Game+ is starting the game over with a fresh character. You dont reuse the powered up character. There are some games like baulders gate and hunters that allow you to resue your powered up character however this is not what he meant. he was talking about games like chrono trigger where you start the game over again. The problem with your idea is that you want to start the game 3 years later after the first play. Thus it becomes a sequal or new chapter, not a replay through the original world with things changed to reflect what you did already. a better idea would be some parrallel stories with mulitple characters. As you go through with each character you get to see some portions of the story you would not normally see playing one of the other charcaters. Wild arms did this. Secret of mana 3 also had some of this as well.

i think the problem with this discussion is that everyone has a different idea of what a persistent world means. here are some of the different definations I have seen used in this discussion as well as others.

1. A game in which the player goes through the same ocntent over again with some changes beased on the players actions through previous playthroughs. Or the New Game+ style. The main game is pretty much teh same, and it will take the player a far shorter time going through because he will know what he needs to do. This in my eyes is not really a persistent world, but is what is being described as what some people want.

2. A game where you go through the game as normal, but towns change as you progress. This has been done in most rpgs after any major event occurs. In final fantasy 4 the netire world becomes destroyed at some point and the world is changed to reflect this. If this is the gameplay you want, then its merely a matter of having such events in the game.

3. You till a certain point in the game where you "finish" the game. You then get to start again however its a later time and your previous actions have affected the world. This is known as a sequal. You simply want to place them as one game instead of spanning mulitple releases. Again no big breakthrough in game design. It was done in shenmue, and shenmue 2. They had i think two chapters each.

4. The MMO persistent world run on a server and consistently being simulated. This is what I would consider a persistent world. however such worlds are only useful for MMO games. However the quests are for individual players and someone else cant slay the dragon before you. This basically boils down into a world consistenting of two parts. The player part in which things can change, but not really affect the world only the history (though WoW is supposed to be adding a way to take over zones through PvP combat). Tne you have the quests which players can do, but dont really affect the world too much.

5. Guildwars does things differently by having all instance quests so the world can be affected by the quests you do. Also the world is shown based on what quests you and all your party members have done so nothing is inconsistent. I place it in its own little bullet because its different then other MMO games in this respect. however its similar to bullet two from above. This game will also have chapters, so it will use elements from type 3.

6. Online persistent world where some quests affect teh world permently and only one person or group can complete them. This requires the developers to constantly be making new content and quests for the players. However it gives the players the ability to affect real change on the world. Wish did this, and it was quite awsome. It also had a player based economy where all items where crafted by players to be sold to shop keepers or other players. There was some "trade" between towns to help regulate the econmy and some key starter items were spawned in the shops so players did not have to worry about not getting the items they needed (however this did not always work out great, and did sometiems leave certain shops empty with items people needed for certain things). Wish got canceled for reasons unkown

7. Player created quests for a MMO game. Face of mankind (in beta fomportal.com) does this. players at a certain rank can create quests for the lower ranked players. This allows for "free" content for the developers, and the players get stuff to do. players can get high ranks through doing the quests, however it is not clear how promotion will work since it does not seem to be fully implemented yet. Also each faction can gain control of colonies thus the story is controled by the players and can affect world changes by taking over colonies. Encomy is also player controlled with no npc vendors. This means the players can control the market as they want with whatever prices they want. Prices were very unreasonable at the start of beta, now people are undercutting others to barely make a profit. So far the game needs a lot of work, and its hard to say if the player created quests will work well. Dependong on the faction you may not have quests because your leaders are not playing or dont play the same time as you.

So what type of system is this discussion about? Personally I see games like Wish being the future of online gaming. Games like Morrowind will be the single player equivilent.

Quote:

Original post by Wavinator I LOVE the idea of persistent, ongoing worlds and the whole "New Game+" concept (forget which game originally had that, read it on the boards here). One problem I've always had, though, is that portion of the RPG which is sheer discovery of the new. If I replay after having a fairly advanced character, I begin to know where everything is. Maybe this isn't so much of a problem, but from replaying Morrowind you can sort of get a taste of this. If the game world expands, you're going to have a memory issue. If the gameworld keeps changing, you're going to have a playtesting, QA and repetition issue. But, of course, that's where creativity comes in. [wink]

Actually I'm not talking about newgame+, but continue+ :). The idea of being able to continue playing the game with a new character from some point in the previous game, whether it be during, or after.

From what I know of your project I think it could work well with a contiune+ system. If you have good solid generated content and a roboust simulation system that makes the game world interesting. Then why not let the player continue playing with a new charcter in the existing universe after the previous game is over? Depending on how good your simulation matrix is you could even let the player choose how many years have passed since their last game ended and their new one begins.

techno goth, i dont think you understand the ramifications of using a simulation system. The keywords "interesting" and "robust" sound great on paper by in reality are very very difficult to implement. If it was worth the effort and gave a great game, it would have been done like that (even if it was partially a gimmick). Generating content based on player actions is one thing. Having a world simulation that would yield new scenarios is quite another.

Questions:

1. What will constitute the end game? Will the player have a certain amount of in game time to play before needing to use a new character (ie age plays a role). If this is the case see Mourning for the pc (a mmorpg about to go through another beta phase). Pirates! also uses age as an end game. Its important to think about this. you could make it that the complete a certain mission instead.

2. When a player continue, he will start after the previous play through. How will you handle simulation of the time elapsed? Such simulation would be very difficult to pull off unless it was some sort of strategy game. Even then, you would have to figure a way to ensure that the cpu does not create a dull landscape for the player. Its quite easy for a single faction to destroy the others leaving the player little to do. A worse scenrio could be that the factions have done nothing to each other since they are in a stalemate so the landscape has not changed much. though that may be a blessing because you could have campaigns where the player joins a faction to tip the balance. Though this wont be too different from some of the other campaign strategy games.

3. Players get attached to their characters and would hate to lose any work put into them. This rules out much of the leveling that you could have in an rpg. This again places you in the strategy genre (not a bad thing if you like those games). However I am sure strategy games have done this to a certain extent already (ie continue+). They create campagins which your actions dictate which ones you will play. It also determines what resources you will have. Unfortinatly strategy games tend not to have the player using a character, instead the player is a group of units.

You could make a great strategy game, but i doubt anything with more complex interactions woudl work. Npcs in an rpg will be difficult to simulate due to the large amount of varibles and more difficult to implement "problems" to occur to warrent a hero. In a space strategy, warring factions could always use help during their power struggle. Simulating who has control is not too bad, neither is simulating a war (since you would use the same ai against the player anyway). However simulating the need for goblins to sack a town for resources is a bit more difficult. Especially if you want it to lead to a larger story about unrest among the goblins against the humans or some other possibility.


Civ might also be a good place to look for inspiration (note i have not played that game too much nor recently, so it may not be too useful to look at).

Quote:

Original post by qazlop techno goth, i dont think you understand the ramifications of using a simulation system. The keywords "interesting" and "robust" sound great on paper by in reality are very very difficult to implement. If it was worth the effort and gave a great game, it would have been done like that (even if it was partially a gimmick). Generating content based on player actions is one thing. Having a world simulation that would yield new scenarios is quite another.

I'm well aware of the difficulties around simulation systems, I wouldn't say it was very very difficult to implement it would be complex problem to solve but once designed it could be implemented relativily painlessly depending upon the design of course. Also just because it has been done yet doesn't mean a great deal there are many advanced gaming concepts that have yet to be done, primarily because they untested and thus major companies aren't prepared to take the risk associated with cost of developing and implementing them in a game without prior evidence that concepts are both worth the effort and will be embraced by the target audiance.

Quote:

1. What will constitute the end game? Will the player have a certain amount of in game time to play before needing to use a new character (ie age plays a role). If this is the case see Mourning for the pc (a mmorpg about to go through another beta phase). Pirates! also uses age as an end game. Its important to think about this. you could make it that the complete a certain mission instead.

It would depend on the design but there would mostly like be several endgame conditions.
1)Main character dies.
2)Certain amount of time has elapsed.
3)Scenario specific goal reached.

Quote:

2. When a player continue, he will start after the previous play through. How will you handle simulation of the time elapsed? Such simulation would be very difficult to pull off unless it was some sort of strategy game. Even then, you would have to figure a way to ensure that the cpu does not create a dull landscape for the player. Its quite easy for a single faction to destroy the others leaving the player little to do. A worse scenrio could be that the factions have done nothing to each other since they are in a stalemate so the landscape has not changed much. though that may be a blessing because you could have campaigns where the player joins a faction to tip the balance. Though this wont be too different from some of the other campaign strategy games.

That would be the crux of the simulation matrix. You would need away to abstract the events that have occured since you last played and then alter the world state based upon the that abstract.

Quote:

3. Players get attached to their characters and would hate to lose any work put into them. This rules out much of the leveling that you could have in an rpg. This again places you in the strategy genre (not a bad thing if you like those games). However I am sure strategy games have done this to a certain extent already (ie continue+). They create campagins which your actions dictate which ones you will play. It also determines what resources you will have. Unfortinatly strategy games tend not to have the player using a character, instead the player is a group of units.

Why does it limit leveling?
Players become attached to their character to the point of replaying the same game with the same character that true. So why wouldn't the opposite work? Playing a new character in the same world?

Quote:

You could make a great strategy game, but i doubt anything with more complex interactions woudl work. Npcs in an rpg will be difficult to simulate due to the large amount of varibles and more difficult to implement "problems" to occur to warrent a hero. In a space strategy, warring factions could always use help during their power struggle. Simulating who has control is not too bad, neither is simulating a war (since you would use the same ai against the player anyway). However simulating the need for goblins to sack a town for resources is a bit more difficult. Especially if you want it to lead to a larger story about unrest among the goblins against the humans or some other possibility.

But the thing is you don't need to simulate indvidual NPCs you need a system of scaleable abstraction, a procedural conversion matrix to alter that game world as appropriate.

A base abstraction for the goblin scenario might look like this:

Global Tier:
Dwarves have begun jyhad against the goblins forcing them out of their underground strongholds.

Country Tier:
Goblins have been forced out of three of five strongholds and causing havoc in the country.

Local Tier:
A Large goblin force is gather supplies in preperation to retake their stronghold.

City Tier:
Goblin attack strength[10] city defence strength[2]
City is sacked.
80% of city resources taken.
20 of 30 people killed.

thats just a rough outline of how it could work. We can always continune this thread a debate on the mechanics behind something like this.