Quote: Original post by Trapper Zoid
Actually, I'm not really making an RPG at all [smile]. At the moment, I'm planning on making a simulated village environment as a test bed for some ideas on character A.I. and automated stories. However, to make it interesting and to ensure it's high quality, I'm going to turn it into a game. It's just that I think it will be a lot more fun if the feel of this simulation was a part homage, part parody of the characters and situations present in console RPGs. It just feels "better" to me, in a way I can't fully explain at this stage (it's still in design!)

This sounds promising. You can google a lot of materials on RPG Parody. You can also get many ideas from MMORPG logs. A parody on MMORPG player-PC personality pairs is an equivalent design.

[Edited by - Estok on July 22, 2005 1:13:55 AM]

Quote: Original post by Estok
This sounds promising. There is a vast deposite of RPG parodies searchable by google (mostly in the form of comics). You may not need to play the games yourself to get the ideas. You can get a lot of ideas also if you play MMORPG and just read the logs. Or alternatively you can make a parody on the PCs controlled by typical MMORPG player-PC personality pairs.

Well, I'm thinking more homage than parody at the moment, it's probably not what you are thinking, sorry [grin]! And although I've only played about half a dozen console RPGs, I've played zero MMORPGs, so I'm not at all qualified to make a parody of that!

The ultimate goal of all this is my dream of creating a game where the computer crafts the storyline to fit the personality and actions of the player (or at the very least, generates a new story from scratch). Player customisability and interactivity are the two things I love most in games, so I want to accentuate those in my games. This village sim. is going to be the first step towards that game (which I hope to finish this decade, fingers crossed!)

However, you've suggested an interesting idea. If I can't find some of the RPGs listed here, I'll probably just download a good quality walkthrough and analyse that. As long as I can get a good quality script of what's happening in the game, that will be useful for research purposes.

Quote: Original post by Trapper Zoid

Quote: Original post by Beige
Final Fantasy 4, 6, 7, 8, and Tactics.

Eight is a bit afield, though.

Chrono Trigger.

Secret of Mana.

Thanks for that, Beige. I was going to look at those games anyway as I've heard good things about them (played Chrono Trigger myself already, absolutely loved it; also played a bit of FF6).

One quick question: why not Final Fantasy 5?

Five and eight are both uncharacteristic; if you're looking to create the stereotypical console RPG, they both have other influences going on. I just put it on the list because it's my favorite game :p

Sort of like trying to learn to draw, and starting with abstract art.

Tactics has a lot of the battle system that Five brought into play anyway.

Oh, also, Grandia and Xenogears. I haven't played Xenosaga so I'm not sure whether to recommend it or not.

Quote: Original post by Trapper Zoid
This village sim. is going to be the first step towards that game (which I hope to finish this decade, fingers crossed!)

From experience, I don't think that a simulation is the first step. It is a lot easier to introduce variability to an existing pathway through components than to create the components that will build the pathway.

Quote: The ultimate goal of all this is my dream of creating a game where the computer crafts the storyline to fit the personality and actions of the player (or at the very least, generates a new story from scratch.

It is easy to get distracted if your goal is big. Instead of exploring each componets individually, I suggest you to completely create a rough version of the ultimate goal from day 1, and then refine the components in the control path.

Think about your goal, it looks like n-dimensional to you. Make a 1-D version of it that encapsulate the gist of the processes. What is the gist?

Climax creation.

Climax can be created by introducing a terrain between the initial state and the final state. For example, the story can be about an evil lord that had come to occupy the village, and the variations of the story resides on how the villagers fight back to expell the evil lord. The important idea here is that the goal is well-defined, and the objective of the game engine is to select/create events that will move the state toward a final state subjected to resource consumption. The distance/resource ratio is the measure of plot's intensity. A climax occurs when the ratio is large (the player had bet everything).

A true-resource is something that limit the number of operations as it reduces. Operations can be characterized by their cost, effect, and risk, where effect is the movement toward the goal state, and risk is proportional to effect/cost. In general, risky options are unavailable unless the resource/distance ratio drops below a threshold.

Your game engine will have an inventory of operations, either scripted or created. If each operation is created on the spot, then the real inventory size is zero. But the component that create/find the operation is really they addressing component for a hypothetical rom that holds all the operations. Your goal is to create the component that can search/create operations matching the current state and especially the distance/resource ratio.

This means that the primary focus is to create a bounded scenario where the distance and resource can be easily quantified, such that the engine can use them to generate appropriate operations.

Lol, Estok you sound like Hannibal Lector when he delivers his "covets" speech.

"What do we do? We covet."

"What is your goal? To create climax."

Eerie...

yea, we are saturated.

Quote: Original post by Estok
From experience, I don't think that a simulation is the first step. It is a lot easier to introduce variability to an existing pathway through components than to create the components that will build the pathway.

Think about your goal, it looks like n-dimensional to you. Make a 1-D version of it that encapsulate the gist of the processes. What is the gist?

Climax creation.

rest of quote snipped for space

If I follow what you are saying, I actually agree with the methodology you outline, however from my experience I'm going to need several components to get the result I want. I've seen the approach you outline been implemented to a degree (well, I think I have, it's hard to follow what you were explaining [smile]), and they are all lacking in character interaction. I've seen models of character interaction, but they are lacking in plot structure. This simulation will have elements of both.

It's really just a tech. demo to get me up to speed on the A.I. side of things, as it's been too long since I've done anything ambitious. I've been thinking about the best way to approach this for a while and I suspect that this approach will be the most rewarding, at least psychologically for me (I suspect I can get this sim. working to a degree, which is something I doubt if I go straight for the ultimate goal on my first attempt).

Anyway, we're drifting from the point of this thread, which is on character archetypes and plot structure in RPGs, and which games are best representative of these. These sorts of discussions are best left to another topic (might I suggest "Automated storytelling and interactive plot in games"?)

reply

If you want to know what makes a good story, why do you look upon RPG as guidance? In my opinion all the Japanese RPGs has crap stories and moronic characters. I think studying Japanese comic is better than studying Japanese RPGs, there are a few outstanding Japanese comic in the last decade, and they have really good stories and characters. An alternative source of study is novels, since authors face tough competitions, all the popular novels must have done the story/character right to beat the competition. Get them and study why they are successful. For example I think Fight Club is quite cool, do you know any Japanese RPG that has such original plots and such interesting character (Tyler Durden)?

Here are some Japanese comics that impress me a lot. I have very high respect for the authors of these stories and I view these comics as classic textbooks to learn how to write good stories and characters.

Slam Dunk (this comic has the most successful character ever found in Japanese comic)
Yuu Yuu Hakusho
Hikaru No Go
Monster
20th Century Boys
One Piece
Eyeshield 21
Rurouni Kenshin
Gantz
Hunter x Hunter
Berserk
Naruto (actually I think the story and characters of Naruto are garbage, but the action scenes in Naruto are exceptionlly well done)

All the comic above do not focus much on romance. For romance, I study these:

My Sassy Girl (Korean movie)
Long Vacation, Tokyo Love Story (Japanese TV show)
First Intimate Encounter (Taiwan romance novel)
I's (Japanese comic)

Quote: Original post by Trapper Zoid
But it's not really just the characters I'm interested in, but also the story structure and plot elements as well.

One shortcut to examine the plot and dialog without playing the games all the way through is to read the script for the game. Gamefaqs.com has most of them.

I'd recommend the following games, all which were published by Squaresoft.

Final Fantasy IV (II in US) -> http://db.gamefaqs.com/console/psx/file/final_fantasy_iv_script.txt

Final Fantasy V -> http://db.gamefaqs.com/console/snes/file/final_fantasy_v_script.txt

Final Fantasy VI (III in US) -> http://db.gamefaqs.com/console/psx/file/final_fantasy_vi_script.txt

Chrono Trigger -> http://db.gamefaqs.com/console/snes/file/chrono_trigger_script.txt

Xenogears -> http://db.gamefaqs.com/console/psx/file/xenogears_script_disc_1.txt and http://db.gamefaqs.com/console/psx/file/xenogears_script_disc_2.txt

[Note: Copy and paste is required since Gamefaq's doesn't support direct linking]