Design a Game - Concept to Polish

Hello again,

This article is also brought over from my website to see that version of this article, follow this link:

On Monday, 4.15.2014, Gregory Avery-Weir of the Charlotte Game Dev Group did a really great talk on how to design a game.  He presented a fantastic step by step approach and broke down into manageable pieces how to design a game from concept to polish.  This article is a summary of his talk such that the world may have access to this extremely valuable knowledge.

In game design, there are 4 major steps.

  1. Concept
  2. Mechanics
  3. Challenge Design
  4. Polish

This process should not be thought of as a waterfall process, this is an agile iterative process which may take several iterations until you have something that resembles a strong game.


The concept is your elevator pitch for your game plus a few rough guidelines.  The exact method in which individuals come up with their concept is different as it is a creative process in which you need to search inside your own thoughts for.  It is extremely important to note that when designing a concept, you should avoid thinking of other games, styles of games etc.  If you use the word “Platformer” or “First Person Shooter”, you will end up duplicating the game play of another game and not necessarily creating a new game.  A concept should be very high level, broad and able to go any direction.

I use the images that appear in my head.

Gregory uses the images that he has in his mind as a starting point.  To demonstrate the creative process of coming up with a concept, Gregory asked the entire group to come up with a few concepts that we might be able to use.  The three concepts we developed were:

  1. Getting past security codes and captchas.
  2. Navigating confusing elevators.
  3. Surviving a black hole.

The group then voted that we wanted to expand on surviving a black hole and thus moved on to phase 2 of game development, Mechanics.


Mechanics is defined as the way in which the player can affect the world through affordances, where affordances is defined simply as the input mechanism in the game.  For example, Mario jumps is a mechanic, while the way in which you use the controller to perform the jumping is the affordance.

So how do we develop mechanics?  First decide what is important to the game.  Typically these are verbs, such as: run, jump, avoid, build, roll, talk.  This is a good time to discuss the scope of your game.  If your verbs indicate “puzzle, solve, push” you might not necessarily want to incorporate “decapitate and fight”.  The mechanics and the concept form the core of your game.

All mechanics should have some sort of consequence associated with them that supports the key concept and core of your game.  If you look at Samus for example, Samus shoots.  In the game she shoots doors to open them.  Everything is about shooting when playing Samus.  In Mario, jumping and hitting a ‘?’ block causes power-ups to come out.  Everything in Mario is about jumping.

So for our surviving a black hole concept, Gregory asked us to pick some mechanics.  We decided on the following mechanics as a group.

  1. Interacting with and entering numbers.
  2. Time becomes faster or slower depending on location.
  3. Navigating hallways almost as a puzzle.

Our affordances for these mechanics would be:

  1. keyboard for number entry.  Mouse or touch for navigating numbers
  2. movement via a directional pad or swiping to create distance from the source of the black hole
  3. same movement via a directional pad.

So the core concept of our game “Surviving a black hole” became: solving number and navigational puzzles involving the stretching and compression of time.  Some key consequences that we decided for the mechanics are:

  1. As you solve numerical puzzles to prove you are in fact human, the space ship robot learns them and can now navigate those passages.
  2. Failing puzzles locks random doors to yourself and the robot until the puzzle is solved again.
  3. You have a time limit to solve all of the puzzles, time moves faster as distance gets closer to the black hole.

So this is pretty cool, we have developed our core concept, the core mechanics a few affordances and some consequences.  This is starting to sound like a pretty cool game.  Now that we have the core of the game designed, its time to develop a few challenges or levels.

Design Challenges

Challenges can be thought of as levels, puzzles etc.  The challenges are the manifestation of complex tasks that you can perform with the concept and mechanics of your game.  The example Gregory presented was the complicated Mario, Goomba, ‘?’ Block system.  The first level of Mario presents most of the mechanics and concepts in Mario and forces you to learn them prior to advancing.  This would be a good first challenge.

Challenge design is an exploration in education.  First you must instruct your player and then you must allow them to demonstrate their mastery.

Gregory hit it on the nose with this and I think it is a fantastic concept.  To have a successful game, you must first present challenges in a way that builds their mastery level of your game.  Start with simple challenges that explores all the cases of your mechanics and progress to significantly more complicated and difficult challenges for those mechanics.

Gregory asked us to design a high level concept for our first challenge, a level 1 if you will.

Our challenge is that the player begins in a room with a single locked door that asks a question “Please prove you are not a robot by entering the code.”  The player must then navigate the room for clues and enter the code.  A giant timer will be at the top and as they navigate the room, it will tick faster or slower.  The player would then navigate outside the room and see a more complex hallway system with moving robots that are stopped by doors.  The player would then have to navigate around solving puzzles to activate the ship so he can survive the black hole.


Polish is what makes or brakes a game.  There are some games out there that are great concepts with great mechanics, but poor polish and failed.  The inverse of this is that there are games with poor concepts, poor mechanics and fantastic polish, but have become great successes.  Gregory laid  out a great list of MUST HAVES for polish.  This list can be thought of as a if you fail to have all of these items, your game is actually missing something.  These items are crucial.

  1. Sound.  Music, ambience, triggered sounds.
  2. Accessibility.  (for colorblind, hearing, what if a player is missing a thumb?)
  3. Skip-able cut-scenes.
  4. Options
  5. Ability to quit or pause quickly
  6. Ability to save.
  7. ability to configure your controls
  8. DO NOT insult the player or waste their time.

Read Complete Rules for Games and The Bill of Player Rights

The 8 items are very important and I can count certain games (Assasin’s Creed, I’m looking at you) that forgot to do simple things, such as enable a player the ability to quit easily, which could have made the game significantly better.

Once you have the necessities of polish, its time to look at what is traditionally thought of as polish, Tuning and “Juice”.


This is the adjustment of critical variables.  These variables can greatly change the entire feel of your game.  These calibrate your game.  For example, think about how long it takes to kill a single enemy or solve a single puzzle or open a door.  What if your game has 1000 doors or enemies and it took 60 seconds for each one.  You now spend half a year opening doors, is that the core to your game?  Here is a list of some items you can adjust.

  1. Walk speed, jump height, acceleration
  2. Audio Volume, Ambience Volume, Triggered Sound Volume
  3. Challenge Length
  4. Font Size, maybe use a scaling algorithm for various resolutions.
  5. Field of View (what you see).
  6. Camera distance, lag and whip-lash.


Juice is all of the extra nice shiny bits that really pushes a game up to the next level.  It is what makes the game seem reactive to what you do, it is what makes specific events appear to be important.  A great example of this would be in Final Fantasy 7, when you defeat an enemy the triumphant “You Win Music” plays, the camera pans around your characters as they do a victory dance in slow motion.  In Candy Crush, things sparkle line up, big explosions and particle effects happen when you get combos etc.  These are all good examples of Juice.  Below  is a list of a few items you can take advantage of.

  1. Particle Effects
  2. Screen Shake
  3. Flash, Fade
  4. Sound Effects
  5. Slow Motion

Juice is fantastic, but just remember that the 100th time somebody watches your slow motion, glittery explosion of juice, they might want to skip it and get on with the game, so though Juice is nice, make sure you keep it under enough control or allow lengthy items to be skipped.


Design is an agile iterative process.  Perform this exercise multiple times on your same concept prior to settling on a design.

Thinking about design intellectually allows me to break down why things are good or bad

When analyzing your game design or adjusting variables, don’t immediately jump to conclusions according to what your players, testers or audience demands.  They may not understand the why of how they feel when they can’t beat your game.  It could be that the challenge is right, you just might need to make your character a fraction of a second faster.  Analyzing your game for rough areas will allow you to determine these things.  Install something like fraps and push record prior to letting testers have at it.  This will allow you to see the manifestation and root cause of the problem.

So the key take aways are:

  1. Concept: strong vision of abstraction for your game, a feeling, a picture.
  2. Mechanics: what verbs do you want your game to embody
  3. Challenges: how can you instruct your players to mastery and then allow them to demonstrate it.
  4. Polish: Must Haves, Tuning, Juice.  This makes or breaks you.
  5. Agile: Do this iteratively, be able to adjust and adapt to the idea.