Go check your handy GDNet Dictionary. The Tree of Death is what a gamewriter experiences when he is asked to script scenes that diverge and diverge and diverge, allowing for dozens of player variables. The number of outcomes quickly becomes unmanageable, even with just a few variables. Sure, there are the traditional smoke and mirrors methods of evasion, bottlenecking outcomes, cul du sacs... even the dreaded "Player asked to make a decision that has no affect whatsoever" syndrome. God, I hate that one. Trickery aside, this is a huge problem. You''ve got all these guys in the forums who''s singular goal is to reach for that golden variable; the ultimate interactivity; complete divergence. I say, fine. Waste your time, and let me play the result if things work out. But damn, how do you plan to script that bad mutha? I mean, voice recordings are completely out, and text will still be an astronomical pain in the ass. Even if you come up with a freaking amazing dialog tree or parser, you''d have to find a freaking bonkers writer to script that mess. On top of that, he''d have to be supra-genious to keep the game from constantly sounding like the worst parts of the first Resident Evil game. Any response guys? Nazrix in particular.

Damn, I hate Microsoft. I typed 2 paragraphs and got an illegal operation which closed the whole window. Anyway...

LF, have you seen the post in the Game Design Forum called Player Actions & Quests? It has another link to the post on gamedeveloper.net which is where the original post was. The original post on wwww.gamdeveloper.net has most of the good stuff, so if you read one then I suggest this one on gamedeveloper

I think it contains some interesing thoughts about allowing player decisions without the tree of death effect. I think the basic key is to allow the player to have certain actions that interact with the world, and allow them to be used throughout the game. That way you don't have to script things as much because the results of the actions will do all the work.

Although, this doesn't exactly provide divergence as much as it provides player choice. I do think this could be a key to divergence though. The hardest part of divergence is giving the player choices. With this method, the game engine would naturally take care of the effects of the different player's choices without the developers' intervention.

Edited by - Nazrix on December 14, 2000 7:06:51 PM

I seem to remember Dwarfsoft was trying to storyboard a divergent story line a little while back. One suggestion was to do it up with HTML and link to resulting events. A good method of keeping orginized, I think.

As far as typing in all the text and coming up with results for every divergent action, I wouldn''t want to touch it as a hobbiest. I haven''t read the Player Actions & Quests thread in awhile, but I think the general idea was to try and plan emmergent properties for a game and to have those be the root plot devises. (Player travels to a restraunt orders food. Food stocks go down. Restraunt orders food from farmer. Farmer goes out into field, dies from wolf attack. Restraunt patrons go hungry. Instant quest.) The major problem with this I suppose would be if you want specific events to happen. If you want events to unfold similar to the Trojan War for example, then I suppose you''d have to weight certain formulas in a certain direction and hope you get similar results often enough.

I'm chasing this white rabbit a bit myself...

I posted in the thread Nazrix is talking about the idea of adopting strategy game conventions for an RPG: Namely, mirroring the way an RTS game's typical base works for an RPG town. Like Nazrix, I too think this will give a lot of choice.

I have often wondered about taking this further. Namely:

Is it possible to look at a story as a KIND of strategy?

I posted about this before but couldn't rouse much interest. Either I explained the concept poorly or it has no merit. I'll try again, tho'

Star Wars, the Strategy Story

Look at some of your favorite stories. I chose Star Wars, not for it's literary merits but because I like it and many know it.

Notice that there are similarities between characters in a story and tokens in a strategy game: Objectives, obstacles, physical locations, characteristics that give capabilities, and so on.

Take Vader. At the start, his objective: Get back the stolen plans. His obstacle, Princess Leia, another token making her own strategic manuevers. We pick up the story at the physical location of Tatooine, where 2 tokens (a Corellian Corvette vs. a Star Destoyer) are fighting it out.

The characteristics that give capabilities is the most important part: This is where you get into personality and motivation, and probably where much of "story" emerges. We could say that Leia's personality drives her to do certain things, associate with certain people, etc, and that this gives her certain options. She sacrifices herself for a greater good. She engages in a bit of deviousness (sending the plans with the droids).

Same with Vader. His personality gives him the capability of killing the captain of the ship, and torturing it's crew. It's vital to see that these are STRATEGIC OPTIONS, and govern the unfolding story as if it were a strategic contest.

So what?
You can vary the story by varying the variables. What if the Corvette had been faster than the destoyer? What if the droids hadn't gotten away? What if, rather than a sacrificial personality, Leia had had a suicidal, or collabratory aspect? What if Vader had had a sympathetic personality, or compassionate one?

How would these changes affect the story? More importantly, notice that they would.

There are three HUGE (related) hurdles to making story, though: Indirect experience of the story, or stuff the player doesn't see firsthand, NPC communication, and authorial control.

All are pretty involved, and this post is running long already, so I'll only talk about these further if there's interest in this concept...



--------------------
Just waiting for the mothership...

Edited by - Wavinator on December 14, 2000 3:49:58 PM

yeah, I admit that is an issue, kseh. I think the more important aspect of the Player Actions & Quests threads is that the player can use combinations of actions to complete a goal. So, specific events could be created, then the player could combine different actions that have cerain rules attached to them to complete the goal.

Like I said, this doesn''t provide divergence directly, but it does provide diverse player choice which I think is a step in the right direction.


"All you touch and all you see is all your life will ever be --Pink Floyd
Quidquid latine dictum sit, altum viditur.
Need help? Well, go FAQ yourself.

Click here to see my current project.

quote: Original post by kseh

The major problem with this I suppose would be if you want specific events to happen. If you want events to unfold similar to the Trojan War for example, then I suppose you''d have to weight certain formulas in a certain direction and hope you get similar results often enough.

Consider the possibility of surrendering authorial control. This is probably blasphemy to died in the wool writers, but RPG game masters do it all the time.

The pleasure then comes in trying to set up an entertaining environment, rather than an artfully woven story. You likely find yourself creating atmosphere, charactacters, and interesting positioning of all of the elements.

But you live with the acceptance that in order to get massive amounts of divergence, some of your work will be wasted (player won''t see it), and the story won''t go exactly as you imagined.




--------------------
Just waiting for the mothership...

Wav,
hmmm...so basically everyone knows that every story has conflict. Every character has motives of some sort. So, each character is trying to attain something encountering conflict along the way. The strategy comes in where each character uses various strategies to attain their goal, and whether they do or not can affect the way the story unfolds. Am I on the right track?


"All you touch and all you see is all your life will ever be --Pink Floyd
Quidquid latine dictum sit, altum viditur.
Need help? Well, go FAQ yourself.

Click here to see my current project.

quote: Original post by Nazrix

Am I on the right track?

Absolutamente!!!!!!

A lot comes down, then, to how tokens pick strategies, and how you resolve conflicts. If you overcome this, and the other hurdles I mentioned, it might be possible to "generate" story.




--------------------
Just waiting for the mothership...

hehe...I don''t think any of this is what LF was hoping for

good ideas and good thinking nonetheless


"All you touch and all you see is all your life will ever be --Pink Floyd
Quidquid latine dictum sit, altum viditur.
Need help? Well, go FAQ yourself.

Click here to see my current project.

quote: Original post by Nazrix

hehe...I don''t think any of this is what LF was hoping for

I think there are only 3 options:

Author conceived content

Player conceived content

Machine conceived content

The first smacks right into the tree at high velocity. THERE IS NO WAY AROUND IT. No method, not secret technique. Highly divergent stories then either become solely collective efforts (teams and teams of unpaid volunteers, good luck) or the domain of corporations who can throw lots of money at the problem (*shudder*).

The second has potential. Look at the Quake or Starcraft community for what player inspiration can get you. The main problem with this, though, is significant loss of authorial control and wildly varying quality.

The last approach is ironically the most far fetched, but the one with the most promise. As author, YOU design the characters, YOU create the world, YOU set the stage... but the machine handles the wealth of combinatorial interactions.

It''s not without it''s many difficulties, however...

quote:
good ideas and good thinking nonetheless

Thx!!!!

--------------------
Just waiting for the mothership...