Game Theory: Level Modelling

I have come to the conclusion that the hardest part of making a game is in modelling the levels.
I can whip up the code for a simple game in under a week (or for a harder game, a month or two), and yet spend the next three months trying to figure out how to make ‘good’ levels for it.

That’s what happened with DEEP Space, I developed all the interesting stuff, and tried making levels, and the project died. Same with Neon Ball. All the coding was done inside of 48 hours, the level\platform sections were make over the subsequent week, and yet I lack the ability to assemble these pieces into a sensible level.

So I’m going to ask two questions:
The first is intangible, and I’m asking it just to see what it generates: what makes a game level fun? Is it challenge? variation? puzzles?

The second cluster may be easier to answer: What is your workflow for building a level? How do you generate ideas for levels that are all in a similar style, and yet different and interesting enough to play? How do you maintain motivation for a single level?

Well, I’m off to get out a pen and paper, and see what I can draw…

I have found a similar problem with my own projects. Core mechanics go quickly. Content (especially levels) creation kills the project shortly after the mechanics are rough drafted. It’s late right now, but tomorrow I can dig through my stuff to see if I saved any level design articles.

I eagerly await to hear people’s response to this post.

EDIT:
I had to wait for a compile to finish anyways, so I dug through my bookmarks. I found a few things, but I don’t recall reading them (probably saved them for later). This means I can’t guarantee any sort of quality or usefulness from these links:

Nice links, Kupoman. I’ll see if i can have time to read those.

There is also http://worldofleveldesign.com/ with articles and guides/ tutorials about creating game levels.

Simple rule to follow: Games exist via challenges, challenges are created by rule sets and those rule sets create limitations. When faced with a goal, the limitation = the challenge and thus over coming the said challenge results in an emotional response we tie to enjoyment.

For level design, one has to be player centric…player centric would imply knowing who your target audience is and where they are starting from. Within a level, you generally want to teach the player basic mechanics via challenges, whether its its a simple jump over a chasm or encountering a simple enemy. You need to introduce them to these challenges but the key is how you escalate the said challenges. Ideally you want to follow the flow chart, which is:

Translation of this pic is as follows: Too much challenge all at once (when they are not guided up to higher challenges via escalation of said challenges) will result in anxiety, which means negative emotional response and thus alienation. If theres no challenge (too easy, no surprise) you will get boredom, which is another form of alienation. Meaning you havent pulled out the emotional response required to immerse the player.

The happy spot is right in the middle and as the challenge goes up, and the complexity with player ability rises, if the player is in the sweet middle spot, they will have the best emotional response.

The other part is presentation and or story. In order for the player to care about their success in the game, you must find ways to pull out an emotional response. Really everything about movie enjoyment and game enjoyment comes right back to the behavioral sciences. Think about all those movies that start off with some tragic event to create the “problem” or challenge the player’s/viewer’s emotions. Imagine starting a game and you are presented with a cute puppy, aww its a puppy… everyone loves puppys…then all of a sudden a giant bird comes down and eats it. “wtf right?” already we have something to not only surprise the player (emotional response) but also something they can relate to (puppys). Now imagine right after that you see more puppys and you show the player more giant birds in the sky. Well obviously the connection is made, without telling the player what to do they interact with the game scenario you present. Their brain is working, birds = bad, puppies = good, goal = prevent birds from getting to puppies. Now you just have to show them how, this can be a gun or any other means to help the player achieve that goal. But wait, whats the challenge that kind of gets in the way? Maybe is ammo count…thus you create a resource management challenge.

Its up to you as the designer to find interesting ways to escalate that challenge while not making it impossible nor by making it too hard right off the bat…escalating challenges. Maybe you start off with 3 puppies and 5 birds and only 10 rounds of ammo. Then you are faced with 5 puppies 15 birds and 20 rounds of ammo…but wait, now maybe there are puppy eating gophers that pop out of the ground. Ok so you are presented with a bat to hit the gophers. Now you have a challenge mechanic that requires the player to identify, process (choose course of action) and react.

Level design pretty much takes the same approach. Challenges are presented in an escalating fashion, visuals should immerse as oppose to alienate the player. Think about what they player must identify and react upon. Know the tricks such as guiding players via light or subtle cues…let them come to their own conclusion without being told what to do or where to go (even if you subconciously guide them) because figuring out on “their own” is what gives a positive emotional response.

For your second issue. You need whats known as a GDD/LDD. Game design document and level design document. The point is that while you might create some mechanics that might be fun, they kind of only exist as a concept that needs to fit within a larger design document. The GDD will create the story/setting, it sets the tone, it describes the game play in great detail from start to finish, it states the target audience and within writing explains everything you need to know about the game and what the player SHOULD experience. The level design doc is broken up per level, how it fits within the larger GDD and has its own set of escalating challenges. The theme will fit the story created within the GDD. Many level designers will block out the levels, either on paper or within a game engine, they will begin to define the challenges and then flesh it out with the help of the environment/prop artist.

Its rarely ever a process of just “winging it”. You have to go in with a clear concept and design, then create the level via blocking volumes and pathways, add in the challenges and then make it look pretty.

Yeah… this is a very interesting topic! I think the most important part of good gameplay is the flow!
What is needed to get a good flow?

Imho you first may not have any problems with the controls. You need to be in full control of the character and know how to control him. The easier it is the better! Secondly the leveldesign should reward you for “good” playing! Say if you need to hit a jump button at a specific point, and you manage to do it right you can get to a specific point in your level, and if you stay in that flow and your timing is good, the level rewards you by jumping on the right places and so on. I hope you get what I mean. It is hard to explain.
There for distances of obstacles need to be in a certain range! Lets say your jump distance has a range from 3-5 units. The obstacle that you need to hit has to be in that area, and if you hit it you can get to another point immediately. Otherwise you have to go a indirect way. If you play at a certain limit range you can achieve certain things easier!
I think that such rewards are very important!

Edit
While writing my post, SaintHaiven posted his article. I think his one is very good!

  1. This is more of a game design question than BGE question and reminds me of a reply I wrote on gamedev a while back:
    http://www.gamedev.net/topic/647341-designing-levels/?view=findpost&p=5090971

You should probably take a look around that site, it has great resources and community to game design.

In addition to that. Levels are the world your game takes place in. It is impossible to have finished game design without having ideas for the levels and what the player has to do in order to advance.

You need to find the desired balance between the aspects like combat and puzzles, spending and restocking, advancing the story and filling the gameplay time, whatever you have on your table for your game.

Perhaps I could say in brief, if your game has a story you need to advance the story every once in a while or if you have a pure puzzle game you have to bring in new elements to that puzzle little by little and start to combine the different elements into harder and more interesting puzzles.

  1. I have pieces of ideas of environments that are linked to different game ideas I have. A windy helicopter platform stirred with dirt. A huge electric billboard that illuminates a tangible space in front of it. An alien shaped glowing chamber repeating in voronoi pattern and shrouding in darkness and mist steaming from the floor. A mazy corridor filled with so many pipes you can hardly see the walls.

After I have enough pieces for a game level I do level layouts on paper thinking about where each piece could go. If there are empty sections I could have ideas about it as I go. I don’t often paint the final pictures on paper, I just make sure I have some ideas about how to style each area as I go. In addition this is the phase where I make sure I don’t have extra rooms without gameplay purpose and that I have rooms for every gameplay purpose I need.

For visualizing what you can think of you will have to work on self improvement because it doesn’t come easy at first. And often you will find you want to make “just” a spooky corridor only as you start trying to draw it you figure out you haven’t really thought about what exactly makes it so spooky.

If you’re willing to share more on your game here or gamedev, I’m happy to help you with it. On general level it is but guessing at best :slight_smile:

What Sainthaven is saying is pretty much on spot. The theory of fun (learning new things is fun), and flow are the first concepts of game design that were being taught in my ludology classes.

As well as what he says about Game design doc. To me, there’s three things every serious game developer should do:
*Write a gamedesign doc.
*Have a good version control/source control system in place.
*Learn programming.

There’s this little webapp called Tiddlywiki which makes making Game Design Docs pretty convenient, as you can throw in pseudo code next to notes about the story and everything will remain fairly organised.

As for level design itself, try to line out what you want to have happen each level. For example, the first level is almost always a tutorial level. So decide upon what mechanics you are going to teach the player, and how you are going to demonstrate what kind of chalanges might await them. Have new challenges happen during each level, or decide upon the challenges existing and divide them up in the levels they may happen in. Let the final level be a combination of all the existing challenges and the most difficult challenges you could come up with.

OK, this is indeed an imperative matter!

For me the most important thing is to identify all final goals!
Player get easily bored, so one must keep the mind and expectations busy.
As one progresses through the level, one must be indeed escalating as the the level approaches the final goal. You then divide it into sections. Each section can then be dedicated to create a specific challenge, and its design must match that challenge/mission, but also, match the mechanics you planned for the game. For that one need to create rewarding opportunities to use such mechanics! Obviously succeeding one level empowers the player to succeed in the next!
Also function can determine form in game design. So the levels should look the way they function!