A truley non-linear story
Author
So I had this idea a while back, and I've fine tuned some of the design details, but I'm not sure if it would be fun to play, or just more frustrating to players. The concept is a sequence driven story. Lets say that in the game there are a couple hundred scripted story events that a player can possibly experience. Each of these events has certain requirements that must be met in order for it to happen (right time/right place, right people still alive, certain events already happened, etc).
Each event will alter the game world in some way (either people living/dieing, attitudes towards the player or other NPC), as a result, the pool of events that can be triggered is made smaller, or directed towards a different path. This would create a web of possible story paths, all being directed towards any number of possible endings, sort of like a gigantic choose your own adventure book.
I also thought that along side this it would enhance the free-roaming feeling to make character death in battle permanent (all except for the main character of course), while making it much harder to die in battle. Having certain characters alive would cause them to either help you or stab you in the back, so their state of affairs would genuinely affect the story line as well.
The question I want to pose is this: would that system make you simply reload the game anytime something seemingly bad happened, or would you let it play out and see how the story unfolds?
it depends... if it is done right i would play to see the story unfold. if however it's done poorly, i would almost surely reload some times. there are too many unclear variables here and you will not get a good answer unless you do a prototype of the game and ask some people to test it. :)
Author
that's the kind of answer I was afraid of. Drat. I can't really explain it in detail without the wall of text problem and I really wanted to get some feedback on the idea before starting to code it.
Are there any examples of games out there like this so I can get an idea for what works and what doesn't? Also, What would make you feel that it was done well? Good writing? A clear understanding of how the gameplay works before you start playing?
Are there any examples of games out there like this so I can get an idea for what works and what doesn't? Also, What would make you feel that it was done well? Good writing? A clear understanding of how the gameplay works before you start playing?
My thoughts:
The player reloads when they perceive that continuation does not lead to a desired goal. The player needs to have a goal and a way to judge whether the goal can be achieve before he can make the decision to reload.
If you don't want the player to reload, you could design the game such that:
1) The player has no goal; or
2) The player can never tell whether a setback is unrecoverable.
Requirement (1) doesn't seem realistic because if the player has no goal, the game isn't a game to the player. So instead of (1), perhaps:
1a) The player does not self-assign any goal
1b) Whenever a goal is made unreacheable, the game introduces a new goal.
The last two ideas suggest a game where the states required to move the game forward is defined for the player. At each moment, the player would be given a number of goals (perhaps only one). The player would attempt to complete any one of them to move the plot forward. If the player makes an unrecoverable mess on one path, the player could try the other paths, the game could give the player a new path to try, or the game could wait until the player asks for a new path.
For this particular design, I am assuming that all possible paths are defined, and that the player is more or less satisfied with the paths the game provides. I also assume that the game world is not an open world where the player is expected to set their own goal.
This is another way to interpret (1) and (2):
1c) The player knows all possible paths to reach the goal, and the player is playing with the mentality to reach the goal by any mean.
In this situation, the player would only reload when all paths are deadend. Because the player has the mentality to use any mean to get to the goal, the player would not be affected by minor setbacks. The player would not care about any setback such as not being able to reach the goal with their pet.
One possible design is a one-for-all design, where, whenever the player passes a stage, the player could resurrect all of the lost members or equipment, such that as long as the player could pass the current stage, there is no loss.
The player reloads when they perceive that continuation does not lead to a desired goal. The player needs to have a goal and a way to judge whether the goal can be achieve before he can make the decision to reload.
If you don't want the player to reload, you could design the game such that:
1) The player has no goal; or
2) The player can never tell whether a setback is unrecoverable.
Requirement (1) doesn't seem realistic because if the player has no goal, the game isn't a game to the player. So instead of (1), perhaps:
1a) The player does not self-assign any goal
1b) Whenever a goal is made unreacheable, the game introduces a new goal.
The last two ideas suggest a game where the states required to move the game forward is defined for the player. At each moment, the player would be given a number of goals (perhaps only one). The player would attempt to complete any one of them to move the plot forward. If the player makes an unrecoverable mess on one path, the player could try the other paths, the game could give the player a new path to try, or the game could wait until the player asks for a new path.
For this particular design, I am assuming that all possible paths are defined, and that the player is more or less satisfied with the paths the game provides. I also assume that the game world is not an open world where the player is expected to set their own goal.
This is another way to interpret (1) and (2):
1c) The player knows all possible paths to reach the goal, and the player is playing with the mentality to reach the goal by any mean.
In this situation, the player would only reload when all paths are deadend. Because the player has the mentality to use any mean to get to the goal, the player would not be affected by minor setbacks. The player would not care about any setback such as not being able to reach the goal with their pet.
One possible design is a one-for-all design, where, whenever the player passes a stage, the player could resurrect all of the lost members or equipment, such that as long as the player could pass the current stage, there is no loss.
Author
Ok, that gives some good food for thought. The thought of the player not being able to tell when something is a setback is something that flows along with the idea I had for this. This was intended to be an RPG type setting, so there aren't any clearly defined stages, but there are clearly defined goals. The "Any means necessary" idea seems really appealing. This I think would also help subtly guide the player down the different paths without making him feel like there is a distinct change in plans.
As far as the goals changing throughout the game, I really like that idea. Since I want this to have multiple possible endings, it would make a lot of sense for the game to present you with the opportunities to take that path.
Example: You fight NPC1, when you get him low on health, he begs for you to spare his life. You kill him. Since he was the only one with information on storming NPC2's castle, the player has cut himself off from that path. As you leave the area, NPC3 comes on screen to tell you that NPC4 might have information to help you get at someone close to NPC2.
Should you chose not to kill NPC1, he tells you the information, and you go storm NPC2's castle, and then go from there.
The idea is to make the player not ever feel like they are in a dynamic environment. I want to make it play smoothly as though every choice the player makes was the right choice. Everything should have that RPG feel of "well something bad happened, how do we fix it". At any point in the game the player will have 3-4 different events that he could go trigger. NPC's will tell the player where and when to go to different areas to start them.
If a PC dies, the NPC's will react to you differently (maybe they were X's friend and not yours, so they won't talk to you now) or when you go to where the event should start, a different one happens because X doesn't piss off Y, or X sees the trap ahead of time, without ever saying "Oh, X didn't come, I guess we'll just leave then" or something jarring like that.
Smooth is the goal I guess is what it all boils down to. Is it possible to make it feel like player isn't just taking a multiple choice test with some fighting in between? Is it possible for the player, knowing that what they choose to do has an impact, to feel like he's playing a linear storyline that fits together, and not just like it's some modular story with mix and match parts?
As far as the goals changing throughout the game, I really like that idea. Since I want this to have multiple possible endings, it would make a lot of sense for the game to present you with the opportunities to take that path.
Example: You fight NPC1, when you get him low on health, he begs for you to spare his life. You kill him. Since he was the only one with information on storming NPC2's castle, the player has cut himself off from that path. As you leave the area, NPC3 comes on screen to tell you that NPC4 might have information to help you get at someone close to NPC2.
Should you chose not to kill NPC1, he tells you the information, and you go storm NPC2's castle, and then go from there.
The idea is to make the player not ever feel like they are in a dynamic environment. I want to make it play smoothly as though every choice the player makes was the right choice. Everything should have that RPG feel of "well something bad happened, how do we fix it". At any point in the game the player will have 3-4 different events that he could go trigger. NPC's will tell the player where and when to go to different areas to start them.
If a PC dies, the NPC's will react to you differently (maybe they were X's friend and not yours, so they won't talk to you now) or when you go to where the event should start, a different one happens because X doesn't piss off Y, or X sees the trap ahead of time, without ever saying "Oh, X didn't come, I guess we'll just leave then" or something jarring like that.
Smooth is the goal I guess is what it all boils down to. Is it possible to make it feel like player isn't just taking a multiple choice test with some fighting in between? Is it possible for the player, knowing that what they choose to do has an impact, to feel like he's playing a linear storyline that fits together, and not just like it's some modular story with mix and match parts?
Interactive storytelling in games is something I was keen on, and hope to get back into sometime soon. One of the first topics I started here was Automated storytelling and interactive plot in games. In the first post there I've listed a collection of resources on the topic that might be of interest. It's a few years old now and I'm out of touch with the area, but I hope they are useful.
Author
After reading through a few of links (some were largely over my head, and I don't really want the story to be dynamically created anyway) I determined that I'm looking for something along the lines of this
I don't like dialogue choices, as those always seem to break up the flow to me. What I was thinking of was more along the lines of the actions a player performs will determine the dialogue choices, so they don't feel like they are picking from a list of outcomes.
in the previous example of choosing to kill or not kill NPC1, we could pose the following situation: NPC1 begs for mercy, you decline to attack him with main character. Main character says "I can't do it, I can't take a life". PC2 pipes up "I can" and deals the killing blow. Now if PC2 had died in the past, he wouldn't be there to to take the action, so after the main character expresses his moral qualm, NPC1 ends up alive. On the other hand, if the player chooses to attack, NPC1 dies, and a little less is known about PC2, but we have a deeper insight into PC1.
Start constraints and end states is exactly what I was thinking. Each story event only starts when certain criteria are met, the player is offered a choice, thought not something super obvious (it could be Fable style where if they player decides to attack the NPC instead of listen/help them then a different end state is produced) and new criteria are set to trigger the next events.
Thinking about this from a programming side as well, it doesn't seem that hard to put a list of starting criteria (NPC1 is alive, player is in City a, Group 1 hates the player), a scripted event, and then give complete control back to the player and have the game wait until an ending criteria is met (player agrees to pay the NPC off, player refuses and leaves, player kills the NPC), where the ending criteria that determines what ending states are set.
[Edited by - EmpyrealHell on December 15, 2008 1:23:55 AM]
I don't like dialogue choices, as those always seem to break up the flow to me. What I was thinking of was more along the lines of the actions a player performs will determine the dialogue choices, so they don't feel like they are picking from a list of outcomes.
in the previous example of choosing to kill or not kill NPC1, we could pose the following situation: NPC1 begs for mercy, you decline to attack him with main character. Main character says "I can't do it, I can't take a life". PC2 pipes up "I can" and deals the killing blow. Now if PC2 had died in the past, he wouldn't be there to to take the action, so after the main character expresses his moral qualm, NPC1 ends up alive. On the other hand, if the player chooses to attack, NPC1 dies, and a little less is known about PC2, but we have a deeper insight into PC1.
Start constraints and end states is exactly what I was thinking. Each story event only starts when certain criteria are met, the player is offered a choice, thought not something super obvious (it could be Fable style where if they player decides to attack the NPC instead of listen/help them then a different end state is produced) and new criteria are set to trigger the next events.
Thinking about this from a programming side as well, it doesn't seem that hard to put a list of starting criteria (NPC1 is alive, player is in City a, Group 1 hates the player), a scripted event, and then give complete control back to the player and have the game wait until an ending criteria is met (player agrees to pay the NPC off, player refuses and leaves, player kills the NPC), where the ending criteria that determines what ending states are set.
[Edited by - EmpyrealHell on December 15, 2008 1:23:55 AM]
I have never tried to create a non linear story or anything, but..
Ever played one of those interactive story books when you were young?
You know, you sit there with a piece of paper, keeping track of your inventory, money etc. And you make choices and the story is driven by those choices aswell as what is in your inventory? I only ever played when I was very young, but roleplaying is very popular among all age groups these days, perhaps they have books for the older generations too?
Eg: A Zombie glowing eyes fix on you from the swamp ahead, and a deep growl that makes the hair on the back of your neck rise, is coming from somewhere in the dark woods behind you.. If you currently hold the silver cloak, proceed to page 15, if you do not, jump to page 45.
Im not sure what those books are called, but you may want to look at how they design and test those story lines. Surely they have ported the concept to pc by now, they may have programs that help you design those stories these days, and algorythms to test all possibilities for flaws.
Ofcourse the problem with those books is the same problem you are mentioning. It was too easy to lie. Too easy to go back and make a different choice, too easy to pretend you do have the cloak, which often resulted in you having no choice but to lie further in the future, because you broke the story lol There is no joy in that, no victory, but you didn't want to have to start at the very beginning either, which is often what happened, as it was easy to end up dead.
Your best bet IMO is to save the story each time a choice arises, and save again each time a choice is made. Some players don't like that, but if you say it won't be so easy for the main player to die, then I see no problem. If the story is compelling and immersive, and interesting enough, then any choice will be interesting to the player. If however, the choices/quests are largely repetitive or the story is the same old unimaginative stuff found in similar games, then it will be boring. Make sure that you aren't basing the story flow on the same variables all the time. You need plenty of different unrelated variables to work with, and a random factor that is out of the players control could be fun too.
Perhaps it is difficult to even get elements of the scrambled story to work in different orders, but make sure that you don't feel victory when you have accomplished that, that just means its working correctly. It does not mean its an immersive story, which is what your actual goal needs to be.
Without a doubt, the biggest challenge is drawing the player into the story as opposed to the player simply wanting to kill stuff. They need to want to know what happens next. Read some books and you will see how they do it.
Ofcourse it is easier to do so with a book. Because a book infact uses only words, and a readers own mind to create a personalized world which is always better for the reader, than someone elses imagined world. That is also why movies almost always fall short of books, for fans. Because the readers imagination is personalized to themselves. Movies and games are limited presentation, whereas imagination is unlimited.
Creating a game instead of an interactive book gives you other benefits though. Complexity being the main advantage. The interactive books often still felt linear because of the simplicity required, because the reader kept track of the variables which meant they needed to keep it simplistic, and work with very few variables.
[Edited by - HumanoidTyphoon on December 16, 2008 4:45:53 AM]
Ever played one of those interactive story books when you were young?
You know, you sit there with a piece of paper, keeping track of your inventory, money etc. And you make choices and the story is driven by those choices aswell as what is in your inventory? I only ever played when I was very young, but roleplaying is very popular among all age groups these days, perhaps they have books for the older generations too?
Eg: A Zombie glowing eyes fix on you from the swamp ahead, and a deep growl that makes the hair on the back of your neck rise, is coming from somewhere in the dark woods behind you.. If you currently hold the silver cloak, proceed to page 15, if you do not, jump to page 45.
Im not sure what those books are called, but you may want to look at how they design and test those story lines. Surely they have ported the concept to pc by now, they may have programs that help you design those stories these days, and algorythms to test all possibilities for flaws.
Ofcourse the problem with those books is the same problem you are mentioning. It was too easy to lie. Too easy to go back and make a different choice, too easy to pretend you do have the cloak, which often resulted in you having no choice but to lie further in the future, because you broke the story lol There is no joy in that, no victory, but you didn't want to have to start at the very beginning either, which is often what happened, as it was easy to end up dead.
Your best bet IMO is to save the story each time a choice arises, and save again each time a choice is made. Some players don't like that, but if you say it won't be so easy for the main player to die, then I see no problem. If the story is compelling and immersive, and interesting enough, then any choice will be interesting to the player. If however, the choices/quests are largely repetitive or the story is the same old unimaginative stuff found in similar games, then it will be boring. Make sure that you aren't basing the story flow on the same variables all the time. You need plenty of different unrelated variables to work with, and a random factor that is out of the players control could be fun too.
Perhaps it is difficult to even get elements of the scrambled story to work in different orders, but make sure that you don't feel victory when you have accomplished that, that just means its working correctly. It does not mean its an immersive story, which is what your actual goal needs to be.
Without a doubt, the biggest challenge is drawing the player into the story as opposed to the player simply wanting to kill stuff. They need to want to know what happens next. Read some books and you will see how they do it.
Ofcourse it is easier to do so with a book. Because a book infact uses only words, and a readers own mind to create a personalized world which is always better for the reader, than someone elses imagined world. That is also why movies almost always fall short of books, for fans. Because the readers imagination is personalized to themselves. Movies and games are limited presentation, whereas imagination is unlimited.
Creating a game instead of an interactive book gives you other benefits though. Complexity being the main advantage. The interactive books often still felt linear because of the simplicity required, because the reader kept track of the variables which meant they needed to keep it simplistic, and work with very few variables.
[Edited by - HumanoidTyphoon on December 16, 2008 4:45:53 AM]
Can't you get the info from the guy, then kill him? Or storm it by brute force instead of info? How about bribing someone within the castle to be a replacement informant? Or getting yourself magically transformed into a friendly looking dog, adopted by a castle inhabitant, and then you would have the run of the place...
In the abstract, I really love the idea of a gameworld that is like a virtual world in that it's not just about killing everything that moves, you can use flattery and seduction and bribery and blackmail and disguises and theft. But to me it destroys the magic of having a virtual world to play in if you don't let the player chose their own goals and especially if it ends up being just another cliche rpg story about war, and because RPG settings are big you won't be able to develop any of it in the depth a virtual world needs. Also, it's not fair to set the plot up like a puzzle to be solved and yet hide all the information about where different actions will lead and still expect the player not to undo actions that had an unexpected unpleasant consequence.
An unrelated problem I have encountered in other non-linear games is that they tend to have interesting events happen too far apart or be missed by the player, with the result of a world both boring and alienating.
But, actually you can see from my posts in that example thread what I think would be the right way to do things. Basically you create a dictionary of objects, characters, places, each of which have some built in basic options and some more options that emerge when the have particular relationships to each other, and yet further option that depend on the pc's current state and what all they have done in the game so far. If you deeply developed a small setting and cast of characters with that kind of structure, you might come up with something really interesting.
In the abstract, I really love the idea of a gameworld that is like a virtual world in that it's not just about killing everything that moves, you can use flattery and seduction and bribery and blackmail and disguises and theft. But to me it destroys the magic of having a virtual world to play in if you don't let the player chose their own goals and especially if it ends up being just another cliche rpg story about war, and because RPG settings are big you won't be able to develop any of it in the depth a virtual world needs. Also, it's not fair to set the plot up like a puzzle to be solved and yet hide all the information about where different actions will lead and still expect the player not to undo actions that had an unexpected unpleasant consequence.
An unrelated problem I have encountered in other non-linear games is that they tend to have interesting events happen too far apart or be missed by the player, with the result of a world both boring and alienating.
But, actually you can see from my posts in that example thread what I think would be the right way to do things. Basically you create a dictionary of objects, characters, places, each of which have some built in basic options and some more options that emerge when the have particular relationships to each other, and yet further option that depend on the pc's current state and what all they have done in the game so far. If you deeply developed a small setting and cast of characters with that kind of structure, you might come up with something really interesting.
Re: EmpyrealHell, regarding to what Sunandshadow posted:
For my own game I am actually using the architecture she is describing, where the game world is system of simple operations (each with an requirement). There is an implementation reason behind this architecture. For me the reasons are emergent events and the lack of scripted events.
You don't need use an architecture like this. It is your choice.
There is nothing intrinsically wrong about a design of a restaurant where the player selects:
0) the house special (aka "you get what the chef happens to cook")
1) the meal
2) the individual dishes
3) the individual components of the dishes (as in a buffet)
4) the ingredients and cooking method (as in the chef cooks the fish you picked)
5) the ingredients (as in pick the fish and cook-it-yourself, cooking utensils provided)
6) the bait (as in you fish yourself, the sea is provided)
You could choose to implement your game at any level (or any mix of them). There is no intrinsic reason to implement it a, just as a diner could be satisfied in a type 0 restaurant where he doesn't even make a choice in the course. Just eat and be happy.
People who make games are in general doing a type 6 activity. So for some of us we could only be satisfied when we have the control of a type 6 situation. It is not a reason why your game needs to be type 6 (high degree of player control), even if it is a non-linear story (which starts at type 2, in the analogy).
For me I wouldn't care if a game only gives me a type 0 situation, because if I want a type 6 situation, I would just go back and code my game. This depends on whether one has the power to fulfill their own desires. I am trying to say that people who enjoys type 0 games are not by any deduction "stupid" or "uncreative" people, and a type 0 game is not by any deduction a "limited" or "stupid" game.
For my own game I am actually using the architecture she is describing, where the game world is system of simple operations (each with an requirement). There is an implementation reason behind this architecture. For me the reasons are emergent events and the lack of scripted events.
You don't need use an architecture like this. It is your choice.
There is nothing intrinsically wrong about a design of a restaurant where the player selects:
0) the house special (aka "you get what the chef happens to cook")
1) the meal
2) the individual dishes
3) the individual components of the dishes (as in a buffet)
4) the ingredients and cooking method (as in the chef cooks the fish you picked)
5) the ingredients (as in pick the fish and cook-it-yourself, cooking utensils provided)
6) the bait (as in you fish yourself, the sea is provided)
You could choose to implement your game at any level (or any mix of them). There is no intrinsic reason to implement it a, just as a diner could be satisfied in a type 0 restaurant where he doesn't even make a choice in the course. Just eat and be happy.
People who make games are in general doing a type 6 activity. So for some of us we could only be satisfied when we have the control of a type 6 situation. It is not a reason why your game needs to be type 6 (high degree of player control), even if it is a non-linear story (which starts at type 2, in the analogy).
For me I wouldn't care if a game only gives me a type 0 situation, because if I want a type 6 situation, I would just go back and code my game. This depends on whether one has the power to fulfill their own desires. I am trying to say that people who enjoys type 0 games are not by any deduction "stupid" or "uncreative" people, and a type 0 game is not by any deduction a "limited" or "stupid" game.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
