Featured Blog

The Scientific Method

The question of "The Art of Games" keeps coming up in discussions. Always present but oft overlooked, however, is the Science of Games. Art or not, these two hypothetical situations demonstrate how many, many games are products and promoters of science.


I seem to die every time I fight this boss. 

How can I get past this boss without dying? 

The boss has a very big machine gun, which fires in long bursts and deals out damage quickly.  He is heavily armored, and as a result, doesn’t seem to move very quickly.  He always seems to know my location, even when I hide somewhere, so he can always track me down.  When hit with a powerful enough blast, such as a grenade, he is dazed for a short period of time. 

I’m focusing too much on racking up lots of damage at once, spending too much time exposed.  By quickly ducking in and out of cover, I can deal out small amounts of damage at a time.  While it may take longer, it should provide the protection I need to outlast my opponent. 

Let’s focus on weapons that can discharge quickly between pieces of cover, preferably while I’m on the move.  Grenades can quickly be tossed.  The light machine gun is great for strafing runs.  A medium or heavy pistol is also in order for quick pick-off shots. 

1. I began by equipping the Light Machine Gun (LMG) while standing in the opening anteroom.  I approached the corner, hugging the wall. 

2. I rushed out to the barricade in the center of the room.  On the way, I fired a quick burst from the LMG.  Having reached the barricade, I immediately ducked down. 

3. The boss proceeded to unleash a long stream of bullets from his machine gun.  I waited several seconds for the wave of bullets to subside. 

4. While waiting, I prepped a grenade for use.  As soon as the firing ended, I quickly rose up and released the grenade towards the boss before ducking back down.  

5. Quickly, I re-equipped the LMG and waited for the blast. 

6. The blast occured, damaging and dazing the boss.  I rose up and quickly emptied a LMG clip into the boss on this occasion.  Before his daze wore off, I ran for third pillar on the right while equipping the pistol. 

7. As the boss recovered, I gazed around pillar and fired three shots.  The boss began firing once again. 

8. I moved to the next pillar, taking a few hits on the way.  Once more, I waited for the enemy wave to subside. 

9. Rounding the pillar, I fired two more shots.  I quickly dove into the second anteroom before the next wave of bullets commenced. 

10. Within the anteroom, I was able to retrieve fresh ammunition for the LMG and Pistol.  Moving to the corner, I waited once again for the wave of bullets to subside. 

11. While waiting, I prepped another grenade, stepping out to release it immediately when clear.  Due to an error in aim, the grenade failed to land as accurately as hoped.  However, the blast radius was still sufficient to damage the boss.  I made a run for the pillars on opposite side of room, re-equipping the LMG. 

12. Following the grenade blast, I proceeded to fire precision shots at the boss character’s head for the duration of his dazed period. 

13. I ran up to the last pillar in the row, equipping the pistol.  Several more shots impacted me along the way as the boss character recovered. 

14. I rounded the pillar and fired two more careful shots.  Following this, I dove back into the first anteroom. 

15. Acquiring additional ammunition in the anteroom, I repeated steps 1 through 7. 

16. Following a final Pistol shot, the battle came to an end. 

Initial results during execution appeared promising.  Though the perceived progress in relation to the time elapsed seemed much lower than in prior attempts, the amount of damage sustained in the process also seemed to be much lower.  While the battle was taking a longer time to complete, it stood a better chance of progressing farther than in pervious attempts. 

This boss has been utterly and completely owned.  In the face.

 I shouldn’t say that.  I barely got through that fight alive.

No, you know what?  Owned.  Destroyed.  Wrecked.  End of story. 

In this instance, panic and haste seem to be less effective means of play than methodical use of cover and speed.  As the boss is based on powerful attacks, it seems to be well-equipped to defend against the same type of attack.  Quickness and a thorough use of defensive cover proved to be the winning solution.  


This puzzle doesn’t seem to make sense to the QA team. 

How can I make this design easier to understand while retaining an adequate level of challenge? 

The existing puzzle is meant to reflect upon the main character’s childhood memories.  As a result, certain childlike elements are incorporated into the puzzle.  A rubber duckie, toy frog, and toy firetruck populate the game area, each one with a different function.  Colored pipes on the floor lead from each object to the next part of the puzzle.  The colors of the pipes correspond to the colors typically associated with each object (red firetruck, yellow duck, green frog).  By activating the objects in the appropriate sequence, players fill in freaky void spaces in the water, swim through air (i.e., fly), and alter the water level to progress through. 

Too many elements are being introduced at once.  If I trim down the number of mechanics used to solve the puzzle, it can help.  By reducing the number of inferences and implied associations, the puzzle should have more universal appeal. 

The idea of twisting corridors has been done to death, but it works well and is easy to understand.  Twisting corridors.  We’re underground in a storm sewer, so water and swimming seem to be logical components.  I have immediate access to assets for wheels, levers, buttons, and lights.  The existing puzzle contains a rubber duck, rubber frog, and toy firetruck, along with null spaces and colored pipes. 

1. Initial re-evaluation.  I examined the nature of the puzzle in relation to the rest of the game.  Which elements are the most vital?  Which are the best fit for the environment? 

2. The existing sewer setting proved important to the story’s progression.  Working from that element, the most necessary feature from an aesthetic standpoint seemed to be water.  Consequently, swimming proved to be the primary mechanic fitting the environment. 

3. Elements of fear are pervasive throughout the game.  A great number of dark, cramped spaces make the player feel crowded and trapped.  In maintaining this atmosphere, the existing concept of a twisting, narrow maze fit in very well. 

4. With these key elements isolated, I proceeded to eliminate the less vital pieces of the puzzle and go back to a more basic starting point – an underwater maze, traversed by swimming. 

5. The basic concept, through fitting in with the game, lacked challenge.  Adding a degree of complexity became the next step.  Originally, this was the function of the toys-and-colored-pipes system.  However, the original system proved too complex. 

6. Re-evaluating the system began with removing its “symbolic” pieces.  This meant looking beyond the specific art assets – the toy duck, frog, and fire engine, along with the element of color – to determine the actual gameplay structure of this system.  That gameplay structure turned out to be a simple one – a sequence puzzle. 

7. Working the sequence puzzle into the maze system meant avoiding the vagaries and implicit nature of the original layout.  Instead of utilizing numerous objects that do numerous things, I began by simplifying the process and having a single type of object do one thing. 

8. Analyzing the gameplay mechanic of swimming suggested that the water itself could prove to be an effective barrier mechanism; the player can only swim as high as the water can reach.  As a result, a secondary mechanic emerged – move the water.  By moving the water, the player can move through the game world. 

9. Thus, the function of the single type of object to use was determined.  When used, this object would alter the water level. 

10. With the mechanic in place, mapping could proceed.  The existing maze from the original puzzle was in place.  Working from its structure, different water levels were worked out that could be used to access different areas. 

11. Upon examining the new mechanics, it occurred to me that the puzzle was not based so much on determining a specific, correct sequence, but on moving from one point to another linearly.  Each object only raised the water level.  All that was required to complete the puzzle was to explore the entire area and locate each object to raise the water.

12. To expand once more upon the mechanic, an element of asymmetry was introduced.  Each object not only raised the water level, but also lowered it in another area.  As a result, some of the locations that could originally be reached with just two moves now required a sequence of five or six moves to reach.  This new layout demanded much more from the player in terms of analysis and problem-solving, increasing its level of challenge.  Still, the concept itself remained quite basic, lacking the implicit nature of the original design.  Incidentally, this mechanic also made more sense aesthetically; water was not simply appearing, but being transferred from one area to another. 

13. From this point, all that remained was the selection of the object to be used to alter the water level.  Several existing options were available to choose from – namely, a wheel to be turned, a lever to be pulled, or a button to be pressed.  The lever seemed to indicate directionality.  Since water was being both raised and lowered at the same time, this directionality conflicted with the gameplay.  The wheel ended up being chosen as it seemed to fit more properly into the underwater environment.  Additionally, it provided greater movement than the button, allowing for a more visible effect upon use. 

Testing the puzzle on my own time seemed to suggest a much more left-brain-centered design than its predecessor.  Even in its early stages, it seemed more logical than the original design.  Despite the loss of artistic flair and symbolic complexity, it somehow seemed more challenging in its new form. 

A new round of testing indicates a much higher completion rate of the new puzzle design.  Reports of frustration and confusion are significantly reduced, as are reports of accidental solving. 

While artistic expression is important to this game from a narrative and aesthetic standpoint, it doesn’t work as well in the logic-based format of the puzzles.  The more artistic and narrative-based puzzle simply seemed to be confusing, whereas its more logical counterpart was more playable and contributed more to the overall quality of the game.  In this particular instance, the story seems to be told most effectively through the atmosphere and the gameplay, rather than through more literary methods such as symbolism and metaphor. 

(No offense intended to the actual designer of “the original puzzle”.) 


How can I tell people about myself without actually meeting them? 

Let’s just say I’ll see you at GDC.

Or rather, I hope YOU will see ME.

Latest Jobs

Manticore Games

San Mateo, California
Senior Software Engineer - Mobile

Sony PlayStation

San Diego, California
Sr. Online Programmer

The Walt Disney Company

Glendale, California
Associate Marketing Manager - Walt Disney Games

Insomniac Games

Burbank, California
Accessibility Design Researcher
More Jobs   


Explore the
Subscribe to
Follow us

Game Developer Job Board

Game Developer Newsletter


Explore the

Game Developer Job Board

Browse open positions across the game industry or recruit new talent for your studio

Subscribe to

Game Developer Newsletter

Get daily Game Developer top stories every morning straight into your inbox

Follow us


Follow us @gamedevdotcom to stay up-to-date with the latest news & insider information about events & more