Designing a game world, to reach 7 secret locations
I'm currently working on a RPG project, quite a traditional one, where the player must retrieve 7 valuable items which have been stolen from a temple.
The items are hidden in 7 different locations among the world, but the plaer won't know that, and needs to discover the locations.
I now face the task to design the world. What I have is the temple location, and a small village close to the temple. The game will start here, and the world will branch off from here. The style of the starting location is fairly colorful with a positivce attitude, a spring/summer style green land.
How to go on? How to decide about the size of the overland and the paths to the different locations?
The locations will have several themes, and the world should gradually change the closer the player gets to any of the locations. So the world needs some size, but I don't want to make it too large. Of course I don't want it too small either. But how to figure a good size?
I'm just starting, so any help is appreciated :)
Are you a programmer, graphics person, designer, or something else? What do you have so far? Project link? Concept art/story? Any other content/ideas?
Need more information like the programming language, framework, design tools, artwork, etc.
I'm a programmer, and self-tought graphics artist. Game design and music/sound are my weakest skills, but usually I get things done there, too. I have a walkaround demo so far: http://sourceforge.net/projects/jewelhunt/
It's java, using LWJGL for OpenGL bindings. No special tools or frameworks.
Not sure why these questions are important to the world design problem?
The story is little more than I told already - The rainbow jewels which give the world the colors hzave been aduced from the temple of the rainbow colors by evil forces of gray. If the player can't retrieve the jewels in time, the world will bleach out and fade to gray.
As said, game design is not my best skill, and that's why I seek for help with that part. I'll also need help with graphics, but I felt it to be too early to ask for that.
I've also got a first version of the raindow temple art:
But actually I didn't want to talk so much about the project, because it's really just starting, and I wanted to discuss only the world design question - given the fact that 7 locations are needed, which the player has to discover, how does one approach world design? How to decide how big is "big enough"? Model a whole continent or just some selected areas, with magic transitions in between?
I ask so many questions to get a better idea of technical restraints, what the platform is, what tools you can (or want) to support, etc. Now I know you're making a 3D world using LWJGL written in Java. That's significantly different from the technical restraints for a platformer using love2d written in lua or a top-down RPG using Qt written in C++. I'm assuming you've rigged your characters in blender. You might do well to get the attention of some of the 3D community members, they probably have experience with 3D game design.
I've been a bit puzzled by the ambush of questions, but well, if it helps :)
But actually I'm not asking a technical question, how to implement such a world. I'm just asking how to create the world itself, the idea, the map layout. Without any relation to programming, just like a classic RPG game master might draw a small continent or big island on paper, with the important places marked and notes for the areas in between.
I have a rough design of the starting location, the temple and the village. The map details for those are not final, but it's sure that these two places exists, and are close to each other.
I have a few ideas about the locations where the jewels will be hidden, but not of all 7 places yet. I've been asking in another forum for suggestions about hiding locations, but the results were meager. But I'll have to start with one location anyways, so I've got stuff to work on. One place will be a classic underground maze, the other will be a tower. I'm pondering about a pyramid and an ice castle as well. The other 3, I have no ideas yet.
The real question is, how to set up the overland area, so that it feels "right" to the player to start travelling from the village and finally reaching one of the secret locations. I wonder how far they should be apart, like, should it be hours of play time between, or rather 10 minutes, or even less? Should I map all the areas, or have magic transitions, like a character who "warps" the player to a secret location as a quest reward?
Technically, it's not a 3D project, it's isometric 2D. OpenGL is just a very efficient rendering engine, which I want to use for color blending effects. Java 2D is lacking in blending modes.
But should be irrelevant to the world design questions.
Edit: I think the idea to paint an island on paper and mark the locations will already answer the question. Lakes, rivers, mountains, forests, jungle, swamp, tribal villages, monster pits etc, I think the ideas will come easily once I actually have started. I just need to take care to keep everything in a reasonable scale to keep it doable.
I may be missing the point here, but you should take a look at Super Mario RPG for ideas about isometric world design. I think that was one of the best isometric platformers I ever played, and it was a hybrid platformer/RPG. Anyway, you should consider moving to a 3D movement system (if you haven't already) to support jumping and other forms of vertical movement.
Syrsly
Twitch Streamer, Web/Game Developer, & Artist
syrsly.com - contact me for commissions, thanks!
Diablo II was my main source of inspiration when I started this. Particularly the item system. But that has to go into another thread, when the project is ripe enough to be announced.
You marked your art assets as isometric, not 3D. I wondered when I was writing my first comment and didn't bother looking, my mistake. You also use PovRay, not Blender, that's an important bit too. The community here tends to use Blender more from what I've seen.
I have limited experience with game design but from what I've seen it's easy to start with a few key areas and develop those well before moving forward. I don't recall where I read it but the starting area is one of the most important parts; that's the player's first impression of your game. It doesn't have to be fancy or have a lot of interaction (think the beginning area for Zelda), it just needs to serve the purpose well. If you begin by designing a whole continent it'll make more work for you later when you need to make changes. Best to start with the starting area, a couple more key points where the player is expected to travel, and the first two or three transition areas between those key locations.
If you have a complicated play system you might start the player out in an area designed as a tutorial and have the option to skip it.
If I were making your game (you have art and code already, that's a good start!) I would begin with the starting area. First major decision is how much information the player needs to understand to begin. Try to keep the learning curve low in the beginning so the player can learn by doing. Nobody wants to read a wall of text before they start playing a game unless it's a graphic novel. One of the important things about games is discovery; we play to find out what happens next. If you start the player off knowning nothing (ignorant peon type hero) it'll be easy to introduce the story one piece at a time.
I look at the original Zelda a lot for game design ideas; it has no real story. It uses pure genious game design to make an addictive experience.
The next place I would design is where the player is expected to find the first crystal. If they need to get equipment somewhere before the crystal I'd start with that instead. Whatever the first major objective is should be your next step. After you have the first two major places and the connecting pieces you'll have a solid idea of how you want things to be designed, what kind of play style to use, the flow of the game and story, etc.
With those things said, I'm no expert with game design. Don't forget one of the most important items in any project; the design document. If you have no goals you can't say when you're done. To the contrary, if you continuously add new ideas and features before they can be finished you'll never be done. Whatever you do, make sure you have a good design document that's clear and allows for measurable progress.
Is this advice helpful?
Yes, that's good advice. Particuarly to start with the key areas and work outwards from those.
The starting areas are particularly tricky, since they should be interesting and good looking to get the player "hooked" to the game, but they must leave room so that other, later regions can be even more impressive.
I also think the first jewel should be fairly easy to get to, to give the player an easy start. Maybe just say the thief dropped it while running away from the temple and it rolled down a drain and likely is somewhere in the sewers now. Sewers are overdone, but this way the player can start adventuring without much travelling.
Side-quest: Get the jewel cleaned afterwards :D
Too much text is a problem. One of my former RPG attempts suffered from that. I thought that people like detailed item stats, but it turned out that there are limits. Thanks for reminding me!
Seems I can start now, the sewers idea will link the first dungeon directly to the starting village, so I can work on a new demo without too much overland design yet.
Thanks, this helped a lot :)
Does this forum have a reward system? I'd like to give you a +1 of sorts for your advice.
There's no reward system for forum posts unless you count happy responses from real people. I appreciate the thought :p
Happy to help in any way. I've only ever completed one game and it's still technically under development. Still polishing an endless runner style game for the Procedural Death Jam (put on by OGA).
I was also inspired by Diablo II very much and that was my first idea for making an RPG. Since joining this site and interacting with the community I've come to enjoy the idea of making a traditional SNES style 2D RPG with some of the graphics here. Good luck and hope to see screen shots and a playable demo in the future.