BBC Article about work with neverending games
Story Here
Dave Mark - President and Lead Designer
Intrinsic Algorithm - "Reducing the world to mathematical equations!"
Looks to me like the classic situation of academia vs the real world.
When the player has too much freedom and little guidance, he may spend more time wondering what to do, rather than actually doing it and having fun. THIS is the first roadblock to free-form games, and it is a design issue. Preventing the player from doing things that don''t make sense is easy in comparison.
I think these guys should look at games like "The Sims" before they draw such grandiose conclusions. Will Wright''s differentiation of "game" and "toy" is crucial in this sense. Games have predefined goals, toys do not. In B&W, the creature part is a toy, while the missions are a game. In RPGs, the character building is the toy, while the story/quests are the game. In GTA3, driving and crashing and stealing is the toy, while missions are the game. Etcetera for sim-type games, RTS games. Purely open-ended games like Elite struggle to gain acceptance from the public. I would expect any serious work on interactive narrative to take this kind of existing grounds into account.
Maybe they do and I missed it.
____________
Jare/Iguana
http://www.iguanademos.com/Jare
When the player has too much freedom and little guidance, he may spend more time wondering what to do, rather than actually doing it and having fun. THIS is the first roadblock to free-form games, and it is a design issue. Preventing the player from doing things that don''t make sense is easy in comparison.
I think these guys should look at games like "The Sims" before they draw such grandiose conclusions. Will Wright''s differentiation of "game" and "toy" is crucial in this sense. Games have predefined goals, toys do not. In B&W, the creature part is a toy, while the missions are a game. In RPGs, the character building is the toy, while the story/quests are the game. In GTA3, driving and crashing and stealing is the toy, while missions are the game. Etcetera for sim-type games, RTS games. Purely open-ended games like Elite struggle to gain acceptance from the public. I would expect any serious work on interactive narrative to take this kind of existing grounds into account.
Maybe they do and I missed it.
____________
Jare/Iguana
http://www.iguanademos.com/Jare
Certainly, open-ended worlds haven''t had much of a go in single player or small group LAN games... but what about MMORPGs... Everquest for example. This sort of narrative engine would be very well suited for such worlds where the open-endedness is currently only provided by employing a team of humans to write new expansions and produce new quests for players to explore.
Furthermore, I think we would see an increase in the number of females playing computer games if open-ended worlds were available. Hack and slash is a predominantly male past-time. Interaction and exploration are certainly things that interest some females and I would expect that the uptake of an open-ended game world among females would be significantly higher than that of linear, level based games.
Just my thoughts though...
Cheers,
Timkin
Furthermore, I think we would see an increase in the number of females playing computer games if open-ended worlds were available. Hack and slash is a predominantly male past-time. Interaction and exploration are certainly things that interest some females and I would expect that the uptake of an open-ended game world among females would be significantly higher than that of linear, level based games.
Just my thoughts though...
Cheers,
Timkin
I think Jare´s distinction between "toy" and "game" holds true, at least with the level of AI we can muster now.
Perhaps one of the best RPG´s ever to be created in my oppinion, Morrowind, was extremely open-ended, but in the end it was in fact the overall plots (ie. game) that made the game worthwhile.
For a game to be never-ending in the way implied in the article, there would have to be an increasing interaction between AI-agents in the game, and they would have to be extremely more complicated than what we´ve seen so far.
Perhaps it would be enough to predifine an abstract goal of sorts to some master-mind main villain (who always elluded the player so he wouldnt be killed - much like some comics, like Batman and Joker) - for instance, world domination - and then have this villain constantly plot ways to achieve this goal.
The villain would have be able to assess and calculate the effect of a vast number of variables in the game that he could manipulate (including other lower-level NPC´s) in order to win (perhaps something like a chess-playing AI). The planning-process could be done on an abstract layer of the main virtual world by a trial-and-error genetic algorithm, and what seems plausible in theory to the villain will then be implemented. But, of course, you, the hero, will be the constant wild-card, the uncalculable variable, that stands in the way of his goals :D (and of course some randomization of variables behaviours in the planning-stage versus the execution-stage).
And when you save the day, the villain starts to plot again, making a new plan, based on the current, altered states of world-variables.
----PATRIOTISM----
“Beware the leader who bangs the drums of war in order to whip the citizenry into a patriotic fervor, for patriotism is indeed a double-edged sword. It both emboldens the blood, just as it narrows the mind. And when the drums of war have reached a fever pitch and the blood boils with hate and the mind has closed, the leader will have no need in seizing the rights of the citizenry. Rather, the citizenry, infused with fear and blinded by patriotism, will offer up all of their rights unto the leader and gladly so. How do I know? For this is what I have done.
And I am Caesar.”
— Julius Caesar
BUSH vs HITLER - who´s your daddy?
Perhaps one of the best RPG´s ever to be created in my oppinion, Morrowind, was extremely open-ended, but in the end it was in fact the overall plots (ie. game) that made the game worthwhile.
For a game to be never-ending in the way implied in the article, there would have to be an increasing interaction between AI-agents in the game, and they would have to be extremely more complicated than what we´ve seen so far.
Perhaps it would be enough to predifine an abstract goal of sorts to some master-mind main villain (who always elluded the player so he wouldnt be killed - much like some comics, like Batman and Joker) - for instance, world domination - and then have this villain constantly plot ways to achieve this goal.
The villain would have be able to assess and calculate the effect of a vast number of variables in the game that he could manipulate (including other lower-level NPC´s) in order to win (perhaps something like a chess-playing AI). The planning-process could be done on an abstract layer of the main virtual world by a trial-and-error genetic algorithm, and what seems plausible in theory to the villain will then be implemented. But, of course, you, the hero, will be the constant wild-card, the uncalculable variable, that stands in the way of his goals :D (and of course some randomization of variables behaviours in the planning-stage versus the execution-stage).
And when you save the day, the villain starts to plot again, making a new plan, based on the current, altered states of world-variables.
----PATRIOTISM----
“Beware the leader who bangs the drums of war in order to whip the citizenry into a patriotic fervor, for patriotism is indeed a double-edged sword. It both emboldens the blood, just as it narrows the mind. And when the drums of war have reached a fever pitch and the blood boils with hate and the mind has closed, the leader will have no need in seizing the rights of the citizenry. Rather, the citizenry, infused with fear and blinded by patriotism, will offer up all of their rights unto the leader and gladly so. How do I know? For this is what I have done.
And I am Caesar.”
— Julius Caesar
BUSH vs HITLER - who´s your daddy?
The point of a this engine is that it can create NPC characters as necessary to enact the narrative, which is dynamically generated by the narrative engine. So, once a villain has been defeated and a story ended, the narrative engine can create a new plot centred around a new villian/subject.
Of course, maintaining an arch-nemesis is a good way to get characters involved in long-term play of the game.
One of the most important things though that I believe open-ended worlds need is the ability to evolve in the absence of the player. In the article above, the narrative is evolved based on the players interaction with the world. It will be necessary to design worlds with AI characters that themselves affect the world. This will free up the narrative engine from creating closed, linear plots that the player sees when they log on each time. Rather, a world populated by intelligent agents would allow the player to log off for a week and come back to a world that has changed dramatically! Of course, this requires that the game world be persistent!
Cheers,
Timkin
Of course, maintaining an arch-nemesis is a good way to get characters involved in long-term play of the game.
One of the most important things though that I believe open-ended worlds need is the ability to evolve in the absence of the player. In the article above, the narrative is evolved based on the players interaction with the world. It will be necessary to design worlds with AI characters that themselves affect the world. This will free up the narrative engine from creating closed, linear plots that the player sees when they log on each time. Rather, a world populated by intelligent agents would allow the player to log off for a week and come back to a world that has changed dramatically! Of course, this requires that the game world be persistent!
Cheers,
Timkin
Which of course also necessitates a Sims-like model of giving the individual NPCs their own desires, goals, tasks, etc. If they aren''t involved currently in the plot, they would be doing other things: working, playing, socializing (which could spread information), etc. That gets very taxing on the system to keep track of all the different NPC''s actions and their effects in an area of the game that may not be revisited for a while - if ever. There almost would need to be an LOD model based on how often the character interacts with those NPCs, how long it has been since he did and the likelyhood that he may interact with them again soon.
Dave Mark - President and Lead Designer
Intrinsic Algorithm - "Reducing the world to mathematical equations!"
Dave Mark - President and Lead Designer
Intrinsic Algorithm - "Reducing the world to mathematical equations!"
Rather than traditional LOD, I have this idea for frequency of interaction to be based on significance to the plot line. So, an NPC adventurer might have a good chance of influencing the plot because of the sorts of actions it can perform, whereas a baker in an out of the way country village has a very low chance of influencing the plot. Indeed, about the only way they CAN influence the plot is if they meet one of the characters or they CHOOSE to influence the plot.
So, given their level of significance (which may change over time as a function of their current actions), update only those NPCs that are most significant. For all of the others, choose their state - upon meeting the player or an important (to the plot) NPC from a probabilistic finite state machine and update accordingly while they are ''in view''.
I know it''s a rather ''loose'' idea... but again, food for thought.
Timkin
So, given their level of significance (which may change over time as a function of their current actions), update only those NPCs that are most significant. For all of the others, choose their state - upon meeting the player or an important (to the plot) NPC from a probabilistic finite state machine and update accordingly while they are ''in view''.
I know it''s a rather ''loose'' idea... but again, food for thought.
Timkin
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement