Building Video Games

Hello World,

Due to my time in the games industry, I still occasionally receive questions around it. Today I received an email from a college student with some questions I feel are more broadly applicable and am therefor writing an article which can be shared to that broader audience.

Question 1: What is necessary to successfully develop a video game?

Honestly, you just need to start building, make decisions and move forward with those decisions.  Often times I see folks get hung up on the best architecture, or the best layout, the best this or that and rework sections of the game so much that the darned thing never gets published.  Decide on your engine/framework what have you, decide on your art direction and just start building.  Get the game to alpha, then refine, then beta, then refine, then publish.  Set solid realistic dates.  “I’m going to build World of Warcraft, mixed with Diablo 3, mixed with Assasin Creed in 6 months” that is an example of unrealistic goal setting.  Understand your abilities, the abilities of your team and assign accordingly.

The thing is game development is software development, there really are no major differences between building video games and line of business applications from a process perspective, its all the same.  Look up anything about agile development and dev ops.  All of that is very relevant in video games and much of the technology used is exactly the same.

Now part of this question is the qualifier “successful”.  The above describes building a game, but perhaps not a successful one.  There are a few key areas here.

  1. Partner with a big organization on some new tech they are pushing.  Microsoft is pushing hololense right now, you better believe there are some companies building games on hololense getting special support and attention.  Find your company, your alignment and show that company the roi and you will reap rewards.
  2. Don’t build the best game you want to play, build the best game the world wants to play.  You are not the world, your opinion does not matter.  A successful game is one in which the world loves it, not you.  Realize you and your team is not the only demographic in the world.  India outnumbers us 5 to 1, what platform do they play on?  What do they like?  What about China?  What about Japan?  Do you know your target market?  Build for them, not you.
  3. Define success and measure against that.

Question 2: What are the stages of game development?

This varies shop to shop, but generally follows this pattern…

  1. Ideation.  Everybody has side projects, ideas, concepts from side work.  Lets prototype a few of those, get some concept art worked up and pick the one with the best chances of filling our definition for success for our next project.
  2. Mockups/Selective Testing.  This phase is all about trying out various flavors of the core idea.  Mechanics, Art etc.  This phase defines the game and direction.  Selective Testing is done side by side to get an idea of response for our target audience.
  3. Build/Test
    1. This is across all departments including user testing.
  4. Build/Test
  5. Build/Test
  6. Repeat steps 3-5
  7. Release Alpha
  8. Repeat steps 3-5
  9. Beta
  10. Repeat Steps 3-5
  11. Test
    1. And user testing.
  12. Test
  13. Test
  14. Repeat steps 11-13
  15. Release
  16. Party
  17. Support Mode

Question 3: How are each of the parts of a game developed?

I’m not entirely sure how to answer this, but I’ll give it a shot.  This really gets back to video games are built just like every other piece of software.  The only real difference is that you have a lot of multi-media and art involved, which is just another department.  The partnership between developers, artists, cinematographers resembles the same sort of partnerships you see between web ui and web back end with dev ops.  They typically sit close together have daily stand up meetings and make sure everything is integrated.  Work streams are always parallelized so there is dev phase, a test/integration phase and a hand off to QA with meetings at each junction at intervals defined by your sprints.  Artists art, coders code, managers manage and the game gets built, there really is no magic or voodoo to it.  Its just standard development with teams that communicate with eachother and ensure their end products integrate.  Exactly like standard software development, just you have a lot more art.

Question 4: What is required to successfully market a video game?

This gets back to question 1 in some ways but I suppose can be expanded upon.

  1. Go to game conferences and bring your game
  2. Social Media like crazy
    1. Automate this as much as possible.
  3. Blog posts
    1. From your artists, developers, managers
  4. Tech Conferences
    1. Talk about the AI algorithms you built, new shader techniques etc.
    2. You are promoting your game to others who then reference your game.
  5. Teach others how to build games, for example the Microsoft retail store, and inform them about your game while you do it.
    1. This is great because you market and teach simultaneously delivering value to the community while promoting your brand.
  6. Compete in Game Jams.

At the end of the day the most effective way to market is to get out there and inform the masses.  This is different depending on your budget and company size, but is generally the same across all organizations just done at different scales.  One thing to be aware of is to ensure your interaction is genuine and consistent.  Also hiring a marketing agency is helpful and worth its weight in gold, literally, they make you money.  Great Game + Zero Sales = no money, dead company, failed game.  Shitty Game + Awesome Sales = Lots of money, live company, successful game; From there you can build your pet project game.


Leave a Reply

Your email address will not be published. Required fields are marked *