Merge open-ended and story-driven games?
Author
I am a developer for Crown and Cutlass (screenshots), a game sort of inspired by the old Pirates! games. We are just getting to the point where we really need to think about gameplay and story lines. I know, if we were doing this right, we should have flushed all this out before really starting to program. Oh well, we're just learning as we go, and hoping to make a playable game! Anyway, I posted this thread on our sourceforge forums. It's kind of lengthy, but I'll sum it up here.
The Pirates! games are a lot of fun, and very open-ended. We enjoy that, but feel like eventually it becomes repetitive. You are just doing the same thing over and over to get more gold, better stats, etc. There isn't much pulling me into the game making me want to keep going. Games like Knights of the Old Republic are very story driven. I enjoy a good story where I actually feel somewhat connected to the characters (more so that just "you're special agent so-and-so, go kill the terrorists!"). However, KotOR is very linear. Games like GTA are story driven and even claim to be sort of open-ended. However, the freedom to do other things is just fluff. You can run around the city, and even do little side missions, but they don't mean anything. The only reason to do them is to get more money, better stats, etc.
Is it possible to merge those two game styles? I would like a game world that is responsive to the players actions. If we do random side-missions (which we probably will do to some degree), I would like their outcome to come into play in the rest of the game.
We are thinking about doing an XML-based mission system, so people can easily add content. I'd like to avoid the feeling like you have this mission you have to do, and the only reason you want to do it is to get to the next mission. I also would like to avoid the feeling like nothing that you do affects the world . I guess I'd like the player to actually influence what happens in the game world.
What do you think? Is that possible? It seems like just about every game shoots for something like this, but ends up falling short. Is this even a good direction to go? What other ways could we keep the player interested in the world of the game? Feel free to post feedback either here or in our sourceforge forums. Thanks for the input.
I'm looking forward to playing this (I think I said that before. [smile]
Open ended games often have a huge problem with variety of encounter and purpose. Story games hook players because of the sense that things are going somewhere, and that somewhere is a meaningful destination.
So what's meaningful about being a pirate? I see two approaches: Either change the world at large (altering prices, port allegiances, vitality of colonies) or impact individual NPCs in a meaningful way (which the story declares happens once you satisfy its trigger conditions).
The first would likely appeal to more strategy minded players, especially if you swing dynamic changes in the world (one intriguing idea for doing this is using influence mapping to change world stats, which in turn govern events that happen to you-- like the chance of encountering a patrol).
The second would probably require a good user community and open scripting, which it sounds like you're doing. I find this hollow in Morrowind, though, because while it claims to change the world it usually doesn't. But maybe if you opened up things like the ability to create new goods on the market, or new ports through fulfilling missions, it could work.
One final thought: By their nature, open-ended games are supposed to be endless, but stories must conclude. I'm not sure how to bridge the gap there (chapters?).
Open ended games often have a huge problem with variety of encounter and purpose. Story games hook players because of the sense that things are going somewhere, and that somewhere is a meaningful destination.
So what's meaningful about being a pirate? I see two approaches: Either change the world at large (altering prices, port allegiances, vitality of colonies) or impact individual NPCs in a meaningful way (which the story declares happens once you satisfy its trigger conditions).
The first would likely appeal to more strategy minded players, especially if you swing dynamic changes in the world (one intriguing idea for doing this is using influence mapping to change world stats, which in turn govern events that happen to you-- like the chance of encountering a patrol).
The second would probably require a good user community and open scripting, which it sounds like you're doing. I find this hollow in Morrowind, though, because while it claims to change the world it usually doesn't. But maybe if you opened up things like the ability to create new goods on the market, or new ports through fulfilling missions, it could work.
One final thought: By their nature, open-ended games are supposed to be endless, but stories must conclude. I'm not sure how to bridge the gap there (chapters?).
Quote:Original post by Mr Grinch
I am a developer for Crown and Cutlass (screenshots), a game sort of inspired by the old Pirates! games. We are just getting to the point where we really need to think about gameplay and story lines. I know, if we were doing this right, we should have flushed all this out before really starting to program. Oh well, we're just learning as we go, and hoping to make a playable game! Anyway, I posted this thread on our sourceforge forums. It's kind of lengthy, but I'll sum it up here.
The Pirates! games are a lot of fun, and very open-ended. We enjoy that, but feel like eventually it becomes repetitive. You are just doing the same thing over and over to get more gold, better stats, etc. There isn't much pulling me into the game making me want to keep going. Games like Knights of the Old Republic are very story driven. I enjoy a good story where I actually feel somewhat connected to the characters (more so that just "you're special agent so-and-so, go kill the terrorists!"). However, KotOR is very linear. Games like GTA are story driven and even claim to be sort of open-ended. However, the freedom to do other things is just fluff. You can run around the city, and even do little side missions, but they don't mean anything. The only reason to do them is to get more money, better stats, etc.
Is it possible to merge those two game styles? I would like a game world that is responsive to the players actions. If we do random side-missions (which we probably will do to some degree), I would like their outcome to come into play in the rest of the game.
We are thinking about doing an XML-based mission system, so people can easily add content. I'd like to avoid the feeling like you have this mission you have to do, and the only reason you want to do it is to get to the next mission. I also would like to avoid the feeling like nothing that you do affects the world . I guess I'd like the player to actually influence what happens in the game world.
What do you think? Is that possible? It seems like just about every game shoots for something like this, but ends up falling short. Is this even a good direction to go? What other ways could we keep the player interested in the world of the game? Feel free to post feedback either here or in our sourceforge forums. Thanks for the input.
The problem with a content system will be the complexities of the game world
that will need to be changeable (how complex do the situations/plots get).
If it is to be highlevel control(easy to change/script) then the precanned (modal) behaviors of the other game entities -- reactions of towns or other ships/captains, placement of objects, etc.. have to be fairly simple. The only aspect that might be flexible would be the text descriptions of the plot (dialogs of standard NPCs, log entries inserted into the players quest log etc...) Most behaviors the player faces will be standard reactions and movements by the other ships. Numbers based abstracts (like crew counts
for battles and treasure payoffs) can be easily handled. Tabular ship stats
and unique crewmember stats would be straight forward.
If it is to be low level (more like programming) then a much more complex interface to the games engine data via a scripting language is needed.
This kind of system could have alot of capability to do logic (if-then situationals and simple AI reactions to different player actions) but would require alot more script conteny to cover the numerous situational factors and contingencies. Im not going as far as programming battle AI, but event keyed sequences that generate muliple plot paths or thresholds to trigger another phase.
Some systems (see Children of the Niles editor) have a wizard driven mission/mapsite editor that can do some multi-claused if-then logic to control the flow of the plot. This flavor is sort of in the middle -- it can do some basic conditional checking and placing of game resources (but is still very clumsy and limited for more complex plot situations).
Its funny, that I actually wish CotN's editor did generate (import/export) XML
because it would be easier to change/tweak much of the scenario data than the clumsy (and unfortunately buggy) wizard style editor they have.
You could try those games too:
- Freelancer
- Privateer 2
- and the best story / open end merge in my oppinion: Privateer 1 (an old, but gold game)
- Freelancer
- Privateer 2
- and the best story / open end merge in my oppinion: Privateer 1 (an old, but gold game)
FFX-2 might be worth checking out - you have a pretty linear main plot line that can be played through in around 10 hours, and is divided into 5 chapters, but in each chapter, there's also something to do in every major location in the game world. Each location's local side-quests, while pure filler from one viewpoint, have a real effect on how that location's plot thread ends up - and giving every thread a happy ending gets you an in-game bonus.
OK, it's not *very* open-ended, but it does give an example of how you can combine linear, world affecting segments into a moderately coherent whole which is, as a result, not entirely linear either.
OK, it's not *very* open-ended, but it does give an example of how you can combine linear, world affecting segments into a moderately coherent whole which is, as a result, not entirely linear either.
I would disagree that there's "no point" to the freeform play in GTA. I think if you DO want open-ended gameplay, having things that don't result in a pre-scripted story sequence is very important. It lets the player make their own story. When people talk about what they did in GTA, their character's story is made up of their own actions: "I took the flamethrower down to the mall and started attacking the people in the Gap, and then the cops came, so I jumped in the car in the display on the main floor and ran over a bunch of them as I crashed through the front door; then the feds we on me, so I had to stop at the paint shop to get the car disguised, but they caught me again when I ran over some people on the bridge, etc. etc" It lets the player determine his own effect on the gameworld in a very free way, rather than having one or two set choices that have been pre-scripted beforehand. I think there's a difference between "nonlinear" and "freeform". As said in the post before me, FFX-2 is a good example of "nonlinear", since you can do almost every quest in any order, but it's not "freeform": no matter what order you do things in, every player is going to basically end up with the same story. You don't have much freedom to actually determine what kind of person your character is or what you want to do; you are given a limited set of set paths and you have to pick one.
Quote:Original post by Mr GrinchI am by no means trying to change the direction of this thread, but I want to take issue with the comments about GTA. Particularly, what I've bold-faced in the above quote.
Games like GTA are story driven and even claim to be sort of open-ended. However, the freedom to do other things is just fluff. You can run around the city, and even do little side missions, but they don't mean anything. The only reason to do them is to get more money, better stats, etc.
The freedom to do other things (side missions), in my opinion, isn't fluff at all, and the very high-level picture for what that freedom "means" is the entertainment value. Perhaps the results of such side missions in GTA don't affect the larger game world as much as a dedicated RPG (in terms of story impact), but after all, as game players/enthusiasts/designers/etc., we often forget the key component of a game is entertainment, while getting extremely bogged down in the details of weaving a story together based on player freedom and choice - the all-elusive "open-ended" design.
Whether or not the GTA series is truly open-ended (even though it's often used as an example of "sandbox" design), the consensus is that it is a very fun, very entertaining game.
I bring up this point of contention with all due respect to Mr Grinch. I suppose I'm just wanting to reiterate the whole issue of entertainment value and what makes something fun. In Mr Grinch's case, he deires that his game Crown and Cutless maintain the interest level of the player, ideally through open-ended design. I agree with him that many developers shoot for this goal, yet fall short somewhere. And perhaps the true, open-ended game has yet to be developed (for many of the reasons already listed by other posters). But if and when it arrives, I'm certain that it will need to be fun to prove successful.
Much thanks for reading,
-Razorguts
I think i'll quote Estok here (since he explained it so well ;D) from a thread Here on Interactive Storytelling.
Estok Wrote:
In contrast to what ishpeck said, my short answer is yes.
There are several properties regarding this topic, including:
- degree of player freedom
- responsiveness of the game world
- diversity of outcomes
- diversity of intermediate plot elements
These are spectrums. It is rather meaningless to discuss trying to achieve infinity in any of these properties (i.e. it is meaningless to say, "I want the player to have COMPLETE freedom")
The simplest form of interactive stories are text-based adventure games. If you think about it, the interactiveness actually declined. There were interactive stories in the beginning, with not just multiple endings, but multiple plots and all the rest. It is only due to the increasing emphasis on gameplay that diluted the complexity of the stories.
Furthermore, any strategy games also fall into this kind of formations. In a strategy games, the plot elements are not presented in dialogues and cutscenes, but in events, such as rushing an enemy base, securing a resource, hiding from the enemy to build an expansion, etc.... (And the player has to perform all of these 'acts').
Interactive stories are games of strategy. In this perspective, designing the mechanism for interactive stories is the same as designing the game elements and AI for a strategy game. In general, these are the things that you need to declare:
1) Forces and objectives - In a strategy game, the objective can be simply to destroy the others.
2) Action freedom - pertains to what units can be built, where they can move.
3) Reaction Rules - this is the overall AI of the forces, such as the building order, what kind of troops to use, what to attack, etc...
4) Perceivable Actions - things that the AI will assess and observe in order to make decisions.
Relating these to the design of a story: (an example)
Estok Wrote:
Quote:Has this concept already been done before?
In contrast to what ishpeck said, my short answer is yes.
There are several properties regarding this topic, including:
- degree of player freedom
- responsiveness of the game world
- diversity of outcomes
- diversity of intermediate plot elements
These are spectrums. It is rather meaningless to discuss trying to achieve infinity in any of these properties (i.e. it is meaningless to say, "I want the player to have COMPLETE freedom")
The simplest form of interactive stories are text-based adventure games. If you think about it, the interactiveness actually declined. There were interactive stories in the beginning, with not just multiple endings, but multiple plots and all the rest. It is only due to the increasing emphasis on gameplay that diluted the complexity of the stories.
Furthermore, any strategy games also fall into this kind of formations. In a strategy games, the plot elements are not presented in dialogues and cutscenes, but in events, such as rushing an enemy base, securing a resource, hiding from the enemy to build an expansion, etc.... (And the player has to perform all of these 'acts').
Interactive stories are games of strategy. In this perspective, designing the mechanism for interactive stories is the same as designing the game elements and AI for a strategy game. In general, these are the things that you need to declare:
1) Forces and objectives - In a strategy game, the objective can be simply to destroy the others.
2) Action freedom - pertains to what units can be built, where they can move.
3) Reaction Rules - this is the overall AI of the forces, such as the building order, what kind of troops to use, what to attack, etc...
4) Perceivable Actions - things that the AI will assess and observe in order to make decisions.
Relating these to the design of a story: (an example)
Quote:Cardinal Prime
This is a multiplot, multi-ending, non-combat, mystery, romantic, interactive game. In this game you plays a jaded professional thief who recently lost the lover. You have been hired to steal an artifact, the Cardinal Prime, from a museum within a week. If the player follows the usual routines of getting prepared, go into the museum, steal it, and turn it in, the default ending is the player being assassinated or that life continues as another day like the day a week ago.
In other words, the time frame is set to be seven days. Within the time frame, the player is invited to make sense out of the situation, by exploring the past, the present, and to make decisions for the future. The player is expected to play the game through multiple game-runs to get a complete interpretation of the story out of many interpretations.
In terms of the 4 elements described above:
1) Forces and objectives
You, your client, your lover. Your given objective is to steal the artifact for your client. However, the story (as well as the ending) will gives hints to you that that is not the actual objective of the game. Your other objectives include finding your lost lover, discovering the identity of your client, the identity of the Cardinal Prime, and your past (there is no amnesia. this is about the one player discovering the past about the PC). Depending on what you have explored and how you interpret the situation, the identity and the objective of your client changes. Your client can range from being the rightful owner of the artifact, to someone taking revenge against you. The objective of your lost lover also changes from helping you to going against you depending on what you do.
2) Action Freedom
This pertains to what you can do within the seven days. The standard actions include: getting equipments, visiting your client, sorting through the stuffs of your dead lover, visit the museum, and to steal the thing. Other options are available as the player declares thoughts, perspectives, and suspicions. Addition actions include: Searching for info about the artifact and the client, sneaking into the client's mansion, visiting your lover's birthplace, visiting an artifact expert, and visiting a weapon dealer.
3) Reaction Rules
This section is large. In general, after you have interacted long enough, the identity and objectives of your client and lover will be set (around day 4). After that they will follow timelines to perform the rest of their deeds. Before that, the events follow a fixed sequence that can be interpreted differently.
4) Perceivable Actions
These are based on the reaction rules. Since I didn't actually show the rules this section is not very meaningful. However, this section includes how when you visit your lover's birthplace she is able to know that you went there. In this game, since the time frame is short, a discrete approach is used to assess the player's perspectives and throughts.
Quote:Games like GTA are story driven and even claim to be sort of open-ended. However, the freedom to do other things is just fluff. You can run around the city, and even do little side missions, but they don't mean anything. The only reason to do them is to get more money, better stats, etc.
I completely disagree. GTA is a sand box game, as open ended as it gets, and with a script-driven story thrown in at the last moment. I had a ton more fun battling it out with the police than I did doing any missions in the game, story related or not. I think I had more fun just cruising around and messing with people than I did playing missions.
A good example of story vs open-ended is Final Fantasy VII. Where the story part starts the game out, then you're turned lose and the world is yours to explore. A lot of people complained that the game became a lot less exciting once the open-ended part took over. My guess was just that the story was so intense. Could have also been the non-story related random battles, or lack of enough interesting locations to explore, or any number of other problems [smile]
The screenshots are looking nice!
I also disagree with the GTA fluff bit - just driving around and exploring / evading the cops could be quite fun.
One other thing, you might want to look into using Lua for the mission system. It's free, easy to use, and can be embedded into many languages - Balders Gate is an example of a commercial game that extensively used it.
I also disagree with the GTA fluff bit - just driving around and exploring / evading the cops could be quite fun.
One other thing, you might want to look into using Lua for the mission system. It's free, easy to use, and can be embedded into many languages - Balders Gate is an example of a commercial game that extensively used it.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
