hi there! i just started a game project, developing an adventure game with a truly deep story. when one looks at present games it might seem impossible but i think there is a chance to make a game that moves the player way more than any movie ever could. computer games are about interaction. thats what make them special, thats what make them what they are - games. in past this point usually was interpreted in the direction that computer games have to make fun in a way that the player is bounded to the game by pressing some buttons as quick an often as possible. i wonder if there is way to effect the players mind with a great story while not boring him with endless dialogues or frustrating puzzles. i really dont know one game that leads the player to deeper thoughts or that impacts a brick of his thick and massive wall of opinions. does interaction harm storytelling that much or is there just a wide empty field to discover and to seed on - where interaction and a moving story support each other?

This is a touchy subject, so be warned.

I, and many ''game'' developers, think that storytelling and interaction have a lot of features that are mutually exclusive. The more of a ''game'' it is, the less of a ''story'' it is. This makes it hard for games to have the same emotional impact of a story.

Other people, usually people from the writing side of things, believe that this is only because we haven''t learned to tell stories in a way that is effective to use in games yet.

The key problem, in my view, is that we don''t have procedural story-generation tools yet, or any real tool that takes a writer''s output and amplifies it. This means that all the plot branches have to be done by hand, meaning that the burden on the writers and the supporting developers goes up exponentially according to the complexity of the story. There''s no practical way of generating the story equivalent of a ''random map'', or of adding in a new tool/weapon/skill to allow the player to experience the story in a different way. The real breakthrough will come when we can create stories without them needing to be written virtually by hand. At the moment, it''s like designing 3D levels a vertex at a time... painstaking and limited.

[ MSVC Fixes | STL | SDL | Game AI | Sockets | C++ Faq Lite | Boost | Asking Questions | Organising code files ]

Jona: sounds like a noble cause. Can you tell us anything about the story, and the way you were thinking of making it happen?

Is the style of the adventure:
1) 1st person slow style like Myst
2) 1st person action style like Outcast
3) 3rd person traditional adventure like Monkey Island or BASS
4) Something else...?

I''m trying to make an adventure game which is heavy in story too. My compromise is to force the player through a story for each chapter, but also allow him/her freedom in the order in which areas are explored, puzzles solved, conversations take place, etc. Some of these things might be completely optional too.

In addition, since the game features five main characters, I''m trying to break the story up into little sub-chapters, each centering around one of the five. (Of course, there may be more than five sub-chapters.) After completing a sub-chapter, the player may choose which character (sub-chapter) they want to play next, thus giving a little more freedom.

wow. interesting replies.
at first let me follow golan and tell you how my story shall work with an interactive adventure game.

when i started the project i wanted the game to be a first-person adventure. my intetion was to get the player closer to the story that way. later my friends convinced me that such an adventure will or might be too boring. (myth was the example they discussed. and i must admit: myth is a nice game but it is indeed boring - maybe because it is just too empty...) so i decided to make the game a monkey island style adventure. i saved the ''get-the-player-closer-to-the-story'' feature by implementing cutscenes in which the main character talks to the real player (that one in front of the monitor).
interaction is about giving the player the chance to choose. my game will run daywise (if that word exists :-)). that means time will go on and on, no matter what the player does. the second main element that gives the required freedom for real interaction are the characters who will be acting completely based on their own intentions. they will do all the normal things any human does day by day - like go shopping, go to work, go to sleep, etc. now, the main character will meet them just when he stays at the same place at the same time as the charakter.
so the whole first part of the game (the second has not been written yet :-)) is only determined by the goal - while the story unfolds itself on the fly.

the first choice the player has to make goes in the direction sage13 described: he/she has to decide whether he/she plays the game in a way to find out all about the story in order solve the quests or if he/she plays the game by just playing around - while the game will be featuring many little subgames and things like that.

i dont know if my system will funtion.
the problem is that a writer wants to tell a story. and he wants the player to at least notice the story, and not only the story but the message behind it. i think the most difficult thing for a writer who writes stories for games is to let loose his aim to effect the player in the precise way he or she wants him/her to be effected.
a game story should never consists only of a single path. maybe this is what kylotan means by his story creation tool. (btw, i really cant imagine such a thing. should it create dialogues to a specified issue or write storylines after the author clicked any buttons? please explain it to me.)

its really tricky. how can i transport my message (or just put any thoughts in the players mind) with a free and developing story?
interaction should take place in a way that the player disusses the problems the story shows with the author, shouldnt it?
but how?

quote: Original post by jona
i dont know if my system will funtion.
the problem is that a writer wants to tell a story. and he wants the player to at least notice the story, and not only the story but the message behind it. i think the most difficult thing for a writer who writes stories for games is to let loose his aim to effect the player in the precise way he or she wants him/her to be effected.
a game story should never consists only of a single path. maybe this is what kylotan means by his story creation tool. (btw, i really cant imagine such a thing. should it create dialogues to a specified issue or write storylines after the author clicked any buttons? please explain it to me.)

I can''t really describe it, because if I could, I probably would have written it by now.

But think about the tools a level designer uses. They can quickly create different shaped rooms, fill them with air, water, lava, attach different textures (organised by type) to the walls, place standard adversaries and objects, and so on. This allows a designer to quickly generate high quality levels, without any significant loss in precision. If they need a unique texture to to alter a single vertex, they can, but usually the standard tools are enough.

Now look at a writer, entrusted with the ''story'' side of the game. What tools do they have? If they''re lucky, they get given a tool that allows them to write branching dialogue quickly. Where are the tools that allow a writer to quickly create a skeleton of a story, and then flesh it out with their own personal touches? Just as a level designer can choose from spheres, tetrahedrons, cubes and so on, maybe a writer could choose comedy, or tragedy, or other basic concepts. Or one of the 36 Dramatic Situations, or something from the Big List of RPG Plots, and have the basics constructed for them. Maybe this would give them some attributes such as the main protagonists and settings which are given default properties that can then be altered by the writer. The system would generate basic conversations or events, and the writer would be left to fill in the exact narrative or dialogue.

Can you see what I''m getting at? The idea isn''t to create something that churns out trash to replace writers, but instead to quickly throw out standard frameworks allowing the writers to concentrate on the details that make it interesting.

The last issue would be how to make this integrate into a game. If the game engine was versatile enough, some of the choices made in the story editor would attach conversation scripts, schedules, AI settings and so on to the characters automatically.

Comments?

[ MSVC Fixes | STL | SDL | Game AI | Sockets | C++ Faq Lite | Boost | Asking Questions | Organising code files ]

Sounds interesting, but I can''t imagine how that would work.

Your first post I was afraid you were talking about a program that would spit out random stories to replace writers, but now that you elaborate, I see what you mean.

I am skeptical that that could work. If there was a real template for every trajedy, or comedy, than stories would all be formulamatic(sp?), and boring. I think comparing it to level design doesnt really work. It would be more fit to compare it to programing. What someone needs to make then, is a new way of writing interactive stories. Like the jump to object-oriented programing allowed more complex programs to be written. What writing needs is simple way to make a story with many, many branches. Thats just my opinion, of course.

PHRICTION

quote: Original post by jona
(myth was the example they discussed. and i must admit: myth is a nice game but it is indeed boring - maybe because it is just too empty...)

You mean Myst, of course. Myth is actually really cool!

Your idea about people going about their day, etc. reminds me of Ultima 7. The adventure game Beneath a Steel Sky also used a thing called the "virtual theatre", where characters walked around all over the map, so their positions were not forgotten once they were no longer on the same screen as the player.

One problem I see with the free passage of time is that you, the writer, lose control of the "tempo" of the main story.

For example, you maybe wanted a big climax where the dragon enters the city, and then the main character goes to the castle to find the traitor before the city is destroyed. With this model however, Day 20 rolls around and the dragon enters the city with no warning. The main character may not even know what''s going on, or that he has to go to the castle.
It''s like one of the Laura Bow mysteries, where you don''t actually have to pay attention to anything, and at the end of the game, you wonder why it finished so quick.

I DO like the idea of the main character talking to the player. That sounds kind of different.

In my game, I expect to have various points where narration will take place, just as speech while you''re playing. eg. as the character enters his room. "my room at the hall is a sty. i never cleaned it up after that tequila night that turned into a scene from fight club... blah blah blah". This just goes on, but it doesn''t interfere with your playing the game.

story writing framework.
ok, i think i understand what you are getting at, kylotan. as writers are creative beings they could have problems with such a thing. one could say that the whole process is too preferenced by that. but i think it''s a good idea and it could indeed be very useful.

at my project we are just programming the graphic engine (btw, we are coding in java) and than we will be doing the game editor. till now we saw it from the programmers state of view, but this discussion make me think the editor could probably be a lot more then just an adventure game creation tool for scene management etc.

so let''s consider what a story creation tool should consist of.
the idea to give it a basic genre is not that bad. i think the important thing here is story points. every story has got a limited set of basic story points. in fact it''s seldom more than four or five, i guess. the rest of the story just grows to or falls from one of that points, if you know what i mean. the tool must provide oportunities to tune on that big story points and the rest of the story should then change automatically accordingly. but what does that mean on detail?
the creation tool can not write a story or part of it all alone. in fact it even cant write simple parts as dialogues or sentences. this still should be in the hands of the writer. so automatically change storylines has to mean: rearrange the relations between objects.
i think that''s what it''s all about: object oriented story writing, as phriction called it.
these objects are characters (of course) and events (in a pretty detailed state of discription). for the tool cant write events by itself, these events has to be described by the author. the point is, that the writer just gives those events names and themes and the creation tool handles them just by that. so the writer could deal with dummy events and could see instantly how one change in climaxes acts on the arrangement of all the little enteties (events and characters). characters, of course, can be such ''big story points'' as well. e.g. one change in the relation between the main character and the evil boss must have a heavy influence on the whole set of things.
wow, as i''m writing this i really get motivated to implement such a story creation tool in our game editor. i''ll give it try for sure. but there are still many things to be discussed.

i must admit that lose of control (or tempo) is big problem with such a free story system that shouldnt be underestimated.
i think there must be some methods that catch this trap. surely it''s not enough just to define the events to happen on day x or y (letting the player fall down with a big question mark over his/her head). but maybe it could work if the events happen somewhen on the timeline defined relatively to other events. e.g. the dragon arrives exactly when the player just comes back to the city. (and if the player dont show any motivation ever go to the castle he has to be motivated by hints and goodies over and over again.)

yes, telling the player a story while he/she does something different is a really good one, too.

i think we are on the right path. (though i dont know where that path leads to :-))