Practical Level Design

February 2, 2021 0 By

Before we get to the main topic of the presentation, let’s go through some common terms in game.

What is a game?
“Problem solving activity, by going through series of meaningful choices, and approached with playful attitude.”
There are many variations of definitions, but let’s keep it simple and easier to understand.

What is a game as a system?
“A balanced relationship between all of the gameplay elements.”

This is easier than said then done, especially now that game has evolved. The more complex a game is, such as in many modern video games, the more elements that it can have. Balancing them all can be very tricky.

Among the common game elements are

  • Rules and mechanics design
  • Goals or objectives (as winning conditions)
  • Game design (level, mission, narrative, system)
  • Interface (physical, visual)
  • Control design
  • etc.

All of those elements, have at least one of these four attributes:

Mechanics – things that concern about the procedures and rules of the game. Deliberate constraints that dictate the gameplay. The very core of what makes the game a game. This is also what makes game different from other non interactive medium, such as movie, book or music.

Aesthetics – things that invokes desirable emotional reaction. It doesn’t always have to be pleasurable, it can be fear, revulsion, or any form of emotion. It is the most apparent attribute that the player can see, hear and feel.

Story – The sequence of events that unfolds in the game. The narrative elements doesn’t have to be as explicit as a traditional story, it can be implicit or as emergence narrative by the way the gameplay is structured, or how the level is designed.

Technology – This is the pen and paper, the deck of card, the dice, the computer system, or any technology that we choose for the game to do certain things and prohibit from doing other things.

These elements have to work together well in a game system, to ensure that the audience, in this case, the player, can have positive experience as they play the game.

In this post, we’re going discuss about an important element which has all these four attributes.


It affects and get affected by mechanics, aesthetics, story and technology.
It can determine whether a video game is good or bad.

What is a game level, really?
Let’s go through some examples.

Here we have the player’s character.
She wants to go from point A to point B and player needs to navigate her through a borderless space. Can this be called a level? Maybe. But is it a good one?
Imagine if this is a really vast space. Can she get lost? Is there any guide for the player to know the right direction?

What if we add a bit of constraint? Now we have a narrow corridor. It is clearer now which way to go to get to point B. But it is so empty. There is nothing the player can do along the way.

Why don’t we introduce a couple of bad guys. Now the player has something to interact with in a form of combat. But is it fun to fight in a narrow corridor?

Let’s widen the space. Now we have a room. Maybe the combat will feel better now. But it’s just a box. It’s not very interesting. And very symmetrical, player may get confused.

Ok, let’s make the shape more interesting. But it is still symmetrical.

Well, add some more walls. Now there are paths for player to choose.

While at it, how about we add more walls for player to take cover. They provide more tactical choices as well.

Adding landmarks would be good. It gives some identity to the room and also help the player to get his bearing.

But still nothing much for the player to do! Ok, add more enemy. Maybe a bit more. A few more. How about a lot more! The more the merrier!

Although maybe not… too much of a good thing is bad enough. Let’s politely remove the excess.

As a change, we can add traps, or puzzle, maybe some platforming challenge for player’s interaction.

It is more balanced now. Or is it? We’ll get to that in a while.

For now, we can understand that game level is basically a space available for the player to accomplish the objective given by the game.

But there are so many combinations to make a level! How do we decide which one is the best? Clearly they are not all equal.

To help us decide which level design is the best, we need to figure out what a good level design is.

Let’s pick these three examples from the previous slide. A corridor, an overly crowded space, and something in between.

If we are considering about a “normal” level, the one in the middle may appear more ideal. It is balanced in term of player’s options, what he can do in the space provided and the challenges given.

But there are times that we may want to have a straight empty corridor like this. One example is when we want to control the pace and intensity. Like what we have here in Metroid Prime. It practically is a corridor. In the game, usually there is at least an enemy in a corridor like this. But before this corridor, player just had to climb up a steep vertical shaft with tricky platforming and some annoying enemies. Can be rather stressful. So this corridor is used as pace control, a breathing space. Additionally, it can be used as loading buffer for the next room.

How about the other extreme? Apparently it is not impossible to have that. In the latest God of War game, right somewhere smack in the middle, they have this. It never happens before in the previous levels of this game, and never happen again in the next levels until the end. It gives this spike of intensity and combined with smart layout and some modifiers, it is challenging, but not impossible for the player. In the end, the player comes out victorious and truly feeling satisfied with the experience.

The take away point is that, there is no magic formula. The level needs to be designed accordingly based on what we want to achieve in the particular space. Do we want it to be serene? Do we want it to be hectic?

A good level design is if it is done based on its purpose and being implemented accordingly.

Ultimately, the level design must highlight the gameplay and support the game pillars.

But how? Good question!

There is no magic formula, yes, but at least, there must be some guidelines!

Luckily for us, many practitioners and academics and even fans have come out with principles, tips and guidelines.

Based on what I learned from people who obviously know much more than myself, combined with what I encountered during my time working as game designer, I like to group the guidelines into the following, basically what we want to achieve with the level.


  • I want to give identity to the level.
  • I want to have a sensible layout for positive gameplay experience.
  • I want the player to have fun activity as we go through the level.
  • Lastly, I want to have smooth continuous flow as we progress through the game.

In order to accomplish that, we need to have a holistic approach.

That means we need to have good understanding of other sub-branch of game design, such as mission design, system design, combat design, narrative design, and mechanic design.

Naturally, good communication with other disciplines is a basic necessity, like engineering team and art team.

Especially if we work in a big team.

Let’s go to the first one. Identity.

  1. The level’s theme, visuals and gameplay must support each other. This is common sense enough. We wouldn’t want to set a game of golf in the middle of a Formula One race track, or war zone, or under the sea. Although… that might be interesting… anyway…
  2. Landmarks! Don’t underestimate the importance of landmark.
    • Landmark can help the player to recognize the room, get his bearing and understand the space. It doesn’t mean that you have to put a gigantic statue or monument in every single room! A landmark can be something simple, yet grabbing your attention. For example a bright red refrigerator on the left side of an otherwise dull colored room.
    • Ideally, unique landmarks offer some form of gameplay, or have an action tied to them. God of War series is usually good at having them. For example the three statues that the player needs to interact with to access the next level.
    • Landmarks can be used to remind player about short term goal, mid term goal and long term goal as well. For example is this looming ominous looking mountain in Uncharted 4 that marks the location of the final level.
  3. Devise your level so that player can identify it with a simple phrase. Like “that room with the waterfall”, “that room with bright red refrigerator”, “that room with three hundred cats.”

Identity can be used as an efficient tool.

  1. The level setup can be used as an approach to fulfill our intention if we know what we want.
    • We can use it to warn or foreshadow an event that will happen in the future. To build up the anticipation, the excitement, or even fear.
      • For example, Tomb Raider puts you in this grisly room with scattered corpses, it warns players that something really evil is lurking nearby. Foreshadowing the boss encounter.
    • We can use it to teach or guide the player. Maybe to use specific ability.
      • For example in Prince of Persia, there are spikes on the floor so you need to do wall running
      • Or half-pipe structure in Metroid  Prime, hinting that you need to use boost ball ability to reach higher ground
    • We can use it to reward the player
      • In Fallout 3 you can find abandoned houses all over the Capital Wasteland. Some of them are falling apart and some are still somewhat intact. By observing it, player more likely finds useful stuffs in the the ones still intact
    • Lastly, we can give greater immersion and entertainment
      • Even if there is no immediate gameplay consequence, by telling a deeper story, we can get the player more invested and care and thus, willing to suspend his disbelieve further, which will help him to enjoy the game more

Speaking of story telling…
Use the level to tell the story!

  1. In games, there are three types of narratives:
    • Explicit narrative – which is well, explicitly called out, like cutscenes, dialog, or written contents
    • Implicit – something that player figures out himself by observing the surrounding and make informed deduction.
      • This technique has been used efficiently by stage shows and movies, there’s a term for that Mis-en-scene.
      • This is the environmental storytelling. We can use the level to implicitly convey nonverbal information. The classic example of this is from Bioshock as Veteran game designer Dan Taylor pointed out in his GDC talk back in 2013. From this one scene near the beginning of the game, we can learn so much about the world. What probably has happened. When things started to go wrong. What kind of people living in this place. It can set the tone and atmosphere for the rest of the game.
    • Emergence story telling
      • This one is commonly occurring in open world game. The various options and choices enable player to create his own version of story.
      • e.g. Hitman. It has various ways to accomplish a mission. Player is given freedom of choice, and by that, freedom to create his own story.

And the three ways are not mutually exclusive. They can coexist together. And together, they can combine their strengths to give a really compelling narrative.

Hitman is again a great example.

  • First, get info for the mission explicitly from the brief.
  • Then player has a lot of options on how to execute his mission.
  • After arriving and observing the guards behavior and the layout, he can decide which path he want to take. What story he wants to write.
  • The final decision is taken after close observation on how the level is being laid out.

Speaking about layout!

Level layout is a great way to enhance the gameplay. A harmony between the two is a potent combination.

So when we do a layout, it should be…

  1. Driven by mechanics
    • Level should be a gameplay delivery system whose primary function is to leverage your mechanics to create great experience.
    • Topology, architecture, objectives, interactions, combat scenarios, etc. should all be designed to showcase your games mechanics. Make sure we have thorough understanding of game mechanics in order to do this.
    • Creative re-use. Not just rinse and repeat. Find a clever way to keep it fresh.
    • Develop gameplay first, fill in walls last — so that you won’t be limited by the walls. And it’ll be more efficient if you need to make changes to the gameplay.
  2. Ideal layout is intimate and bespoke — it means that the pacing and density of gameplay and traversable space is intentional.
    • Every part of the room should have a purpose — don’t build a box and try to fill it.
    • Room size, be it small or large, there is an immediate gameplay reason for it
    • Give the player reasons to traverse the space fully — along the sides, through the middle, under or over something, etc
    • Different room shapes are appropriate for different gameplay scenarios. Not every room should be a cylinder or a box.
    • Simple things can be meaningful choices — even steering left and right
  3. Avoid pure symmetry – this can be disorienting.
    • Even subtle changes to a symmetrical layout can help. For example, changing the heights of doors or floors. Or by placing the the bright red refrigerator on the left side, instead in the center.
    • Be careful with perfectly cylindrical space
  4. The moment a player enters a room is your chance to make a first impression.
    • Think of what you want the player to see and understand right away – this will inform the player of what choices to make.
  5. Normally in games we don’t give player all abilities from the start. Design with future abilities in mind so they can be experienced in new and different ways during the re-traversal.
    • It can be very explicit, like placing a collectible on previously unreachable area.
    • Or it can be as simple as making it more efficient to navigate through the level. Even this can make the experience fresh as player is trying to get a new objective.

  1. Clear objectives
    • Keep guidance concise and clear. On the other hand, only provide guidance by giving as much information as absolutely necessary.
    • Tell players what to do, but not how to do it. Introduce the problem, let player work out the solution. One of the worst thing to do is if a player accidentally solve the puzzle without realizing it.
    • If there are multiple ways to accomplish the goal, that is not bad, but we need to be careful because it may give opposite effect. Too many choices can can risk to confuse the players easily.
  2. Fun to navigate…
    • However challenging a level is, we want the player to enjoy it. Even if we intend the player to struggle and get frustrated, it should still be frustratingly fun to do. If that makes sense.
    • When player go through a level, he will observe, then strategize, then navigate. It can happen slowly, one after another. But all three can also happen in a split second.
  3. Consistent visual language will go a long way to help player understand the rule and the space.
    • Make the level intuitive to traverse. Use consistent metrics. If a platform looks like it can be jumpable, it should be. Otherwise, it will confuse the player.
    • Make clear distinction on where the player can go and where he cannot, and what activity can be done in a particular pace.

Which brings us to the next slide about activity.

There should be enough things for player to do in the level to keep it exciting.

  1. Maybe it can be as simple as doing an observation.
  2. Or navigating oneself through the environment.
  3. Solving puzzle will provide good brain challenge for the player. If done right, it is a very powerful tool.
  4. Interacting with objects, operating equipments, collecting collectibles, talk with NPC or AI
  5. But let’s focus a bit more on combat
    • Try not to give generic consideration to the idea of ‘combat.’
    • Enemies are either part of the experience of traversing a room, or exist as an intentionally sequenced encounter.
      • Example of enemy as part of navigation experience: in Mario games, we often find goombas specifically placed as moving hazard, on a path or platform that player has to go through.
        The placement is deliberate. It makes the player think more about how they will jump or navigate.
      • Enemies as intentionally sequenced encounter: in The Force Unleashed II, in a part of Cato Neimoidea Western Arch level, we had 6 different types of enemy to place in here. In a game where one of the pillars is going crazy with the Force, the main character is naturally over powered. But if player can just go through wave upon wave of enemies like hot knife through butter, it will get boring very quickly.
        So, on top of setting up waves of different types of enemies with varying degrees of difficulties, I also set a group of Scout Troopers, the sniper class, to periodically target the player from long range. Every time player is making a progress, they will retreat to safer distance beyond reach, and finding chance to snipe the player again.With this, I tried to achieve three objectives:
        • First, to incentivize player to use specific mechanic, the force dash, to avoid sniper fire
        • Second, to give meaningful challenge to the player, instead of just mindlessly mashing the button. The player may have stronger motivation to keep on pressing forward and feel satisfied as he catch up with the Scout Troopers.
        • Third, it gives an illusion that the enemy is strategizing and coordinating their attacks, thus it may give additional layer of believability.
      • Too much of the same type of enemy (or same kind of complexity) in repeated sequences gets tedious. Have variations. But remember to keep it under control, we want to challenge the player, but we also want them to feel victorious.

If player can enjoy going through the level, the progression will feel fun and rewarding.

Speaking about progression…

To ensure that the gameflow feels natural and not forced…

  1. Think of not only your room but how it relates to those on either side of it.
    • What is the pacing of the zone? When does the player use certain abilities or gain specific items? When are there combat, puzzle, or exploration? Pre plan all of this ahead.
  2. Constantly teach player something new
    • By introducing new mechanics we can prolong the enjoyment.
    • New ways to learn new skills, new equipments.
    • By periodically flipping players out of the comfort zone, we can create a lot of fun.
    • Learn play challenge master surprise
  3. Keep it fresh and surprising
    • Doesn’t mean to set a lot of jump scares.
    • Require players to use the old equipments in unusual ways.
  4. Empower the player as they progress. In other word, make them feel good! Make them proud and satisfied after beating the enemy. Or make player feel smart every time they can solve a puzzle. Don’t hold their hands too much. Give challenge, but don’t make it frustrating. Give leeway, but don’t make it too easy.

And those are the guidelines.

Having said all that,

  1. The rules are not set in stone. Game development is not pure science, nor it is a fine arts. It can be highly subjective. From time to time rules can be broken, in some cases it has to be broken, in the service of a new idea or fun opportunity.
  2. Take risk, but plan well. Design diligently on paper, but don’t expect that it will always work after implementing it. Get to greybox it, prototype it and test it as early as possible.
  3. Listen to anyone’s feedback, but also listen to your intuition. Pay attention to even minor instincts and feelings when playtesting your room. Be critical and be honest. Don’t get too attached to an idea.
  4. Ultimately, we don’t make games for ourselves, we make it for the players to enjoy. So know who they are and understand what they want.

One thing for sure, as game developers, we are expected to do the unexpected.

And that is the end of today’s presentation.

Thank you very much!