This site showcases the thesis capstone projects for the Full Sail Mobile Gaming Master of Science program. Students completing the program post their end of program project self evaluation here examining what went right and what went wrong during production.

The site provides examples of all completed projects, without regard to the quality of work. Final faculty evaluation of your project is separate from your postmortem. It is a place to share student work and start dialogue with faculty about completed and upcoming projects.

If you are adding a postmortem for a completed project to this blog, please do your best to provide a meaningful meta-level evaluation of your project. This helps students currently in the program have a better understanding of the critical points related to independent production, game development and design and project management. The template for the blog content and instructions can be found in the first post from July 2014.

Thank You,
MGMS Faculty

Sunday, October 23, 2016

Defend Thy Castle

Capstone Game Post Mortem: Defend Thy Castle

Brian Welch

MGMS Program

Full Sail University

October 23, 2016

Defend Thy Castle

Real Time Strategy


Pay for download $0.99

Unity 3D, C#, Javascript, Source Tree

My intended audience is 15-25 year old Real Time Strategy fans who are looking for a new approach to RTS games


Defend Thy Castle  © 2016 Gaming Chef LTD

Command the Impossible

Defend Thy Castle is a real time strategy game that places the user in-charge of a flying castle on a quest for world domination.  The user is able to deploy troops to defend the walls, aim cannons to combat overwhelming waves of enemies, and cast powers to even the odds.  The user must do whatever it takes to survive and conquer the ten levels. Defend Thy Castle’s unique selling point is that it takes the RTS genre and mobilizes it by making the user’s base fly towards the enemy base creating a more dynamic setting for the battlefield. 

Demo Video:
My original concept was to create a hybrid time management and strategy game drawing my inspiration heavily from Kitchen Scramble and Diner Dash in terms of incorporating the time management game play into a battlefield context.  The final version resembles a game that fully favors RTS game play in the same way Warcraft would with the exception that the user’s base is flying toward the enemy base and is not static.

The scope of this project was to achieve developmental mastery through the use of artificial intelligence.  This concept has been established through the use of an enemy algorithm to analyze and select optimal attack waves based on line of sight and user defense information.  Once an attack selection is made, this algorithm is further augmented with a path-finding algorithm attached to the enemy units as they are deployed onto the map to attack the approaching Castleship.

I was hoping for 30 levels divided into three zones to present along with the developmental mastery. This addition can be added through more research on the Asset Store for terrain assets.  If the Asset Store is insufficient, a freelancer can be hired to supply the actual artwork before the game is released.  The rest of the mechanics have gone according to plan.

The Critique: What went right…

The three dimensional graphics that I sourced from the Asset Store have largely been a success as they are high quality without having too high of a polygon count to drag down my frame rate.  In addition, the particle effects for the explosions and lightning have also been given good feedback through user testing.
The three primary user interfaces went well. The first interface is what allows the user to deploy troops to a wall and this is done through a radial interface that appears when a user touches the wall and allows the user to drag their finger to select a troop to deploy to that particular wall.  This functionality is pictured below.
The second portion of the user interface that went well was the cannon reticle UI controls that allow the user to drag and aim each of the Castleship’s cannons.  This concept evolved throughout the course of the project and has been the most frequently used UI control throughout testing.  Pictured below the cursor represents finger placement as the cannon’s reticle is dragged towards the oncoming enemy waves.  Notice the cannon rotates behind the cursor to follow the trajectory path. 
The final portion of the UI that was successful was the special character spells.  These are represented by the three icons floating around the main character at the center of the ship and are activated by simply tapping and dragging to any wall of choice.  Two of the powers are uniform throughout each level (heal and repair) and the top power is changed every map.  This particular picture shows ice as the selected power.
The GANNT chart Milestone management through Underdog helped facilitate productivity and the core functionality has been completed.  Project Management is covered further in the below ‘what went wrong’ section.

Unity 3D has been efficient and enjoyable to work worth.  I utilized the new Unity GUI system heavily throughout my game and I feel that this Engine specific feature really helped facilitate the effectiveness of my overall user enjoyment and interface.  The Pathfinding Algorithm has gone tremendously well with my enemies fully able to respond to the user’s Castleship as a dynamic obstacle.  Below the Debug mode draws pink lines to show the current paths each enemy takes to their respecitve assigned lane of attack.  Please note that there is an ‘I’ collider over the user’s Castleship as a dynamic obstacle to ensure that the enemies only approach from their assigned lanes.

In addition, the Enemy AI Analysis algorithm successfully cycles through each lane and is able to dynamically determine the user’s defenses based on the line of sight of the enemies on the screen.  In the below picture the LOS of the enemy is represented by the green collider and the eyeball is the notation that the AI uses to track which lanes can be analyzed.  From this information the AI is able to determine the best wall to attack.
A good portion of testing was informal and done to see how natural/intuitive the UI controls were.  This testing has led to the current UI system as the troop deployment UI has been drastically changed several times throughout this development cycle.  I feel that the final UI has been a direct result of the testing done to ensure that the interface controls were user friendly.

Once the new levels have been completed, I have decided that my game is not suitable for interstitial ads and have decided to opt for pay for download option with the price of $0.99.   My marketing plan will consist of the following elements when I release my game to Google Play:
  • App Store Optimization (ASO)
  • Google Adword Campaign
  • Press Releases
  • Game Reviews from Sites and Blogs
  • Youtube Influencer reviews
  • Forum Posting
  • Social Media presence – Twitter , Facebook, Pinterest, Stumble Upon, Youtube, and G+

The two main elements of my Aesthetics that have fallen short are the cloud background and the general center of the ship.  As seen in the below picture, the clouds are a bit blurry and are often mistaken for static waves and the center of the ship only reinforces this obscurity.  These elements will have to be revisited before the game is actually released. 

I need to, also, design 20 more levels to organize into three zones.  Since, I will be charging $0.99 for the game I will need to ensure that there is at least an acceptable base amount of content to justify the expenditure. 
My project management has suffered many delays.  I had to do two Green Light Presentations, I was hospitalized and set back two months, and I had to repeat one month in my development phase due to a complete lack of progress.  I need to be better organized and have a better handle on my time.  The GANNT chart was helpful for keeping me accountable, but it still was unable to break me of my last minute tendencies.

There could be more choice given to the user for defending the actual walls.  This point would also help show how dynamic the analysis algorithm is because it could successfully show how the computer is able to read and respond to the current defensive situation on the user’s wall.  This issue can be resolved with the addition of new units.  However, the models will have to be of the same quality of my pre-existing units and it will be difficult to add more than 1 unit with current radial selection UI. 

X  X
I need to have more focused testing on my graphics as I feel that my background graphics have let me down.  This testing will be done through a couple rounds of A/B Testing through Preapps and Facebook, which will hopefully give me a better solution. 

I have changed the business model to a paid game instead of a free game because I didn’t feel that ads were suitable for my game. In addition, because I currently have no in-game currency for promotional video ads or any type of in-app purchases to offer, the paid direction seems the most economical. 

                  Though my project has evolved significantly from the original concept, I feel all the changes have been beneficial and have massively helped improve the overall project.  The main trade-off was switching from time management/ strategy to real time strategy as this changed my core game play focus.  From this experience, I have learned the importance of articulating my ideas from the beginning of the project and, also, the importance of doing thorough research on my assets before I begin coding.  However, if I had to do the project again, I wouldn’t change anything, as each milestone was an essential part of my learning curve. 
                  I am currently paving the way for this game to be released in January of next year with several other game releases to help establish my brand and market presence before Defend Thy Castle is released.  In addition, I will be using the revenue generated from these other games to go with a reputable marketing agency to help promote my game correctly from the onset.  I have already experienced how quickly a poorly marketed game can end up in the app graveyard and I will not be making these mistakes for Defend Thy Castle. 

SourceTree. (n.d.). A free Git & Mercurial client for Windows or Mac. [OSX Application]. http://www.sourcetreeapp.com/

Unity. (n.d.) Unity – Game Engine. [Android Application]. http://unity3d.com/unity/download\

Welch, B. (2016). Defend Thy Castle [Android Application]

Big Dragon
Cannon Design
Castle Design
Character Design
Level Design
Small Dragon