Archive of TideArt content.

Mon, Jul 25 2011 20:09:12 UTC

Interview with Tom Chiu, level designer

Today we have an interview with Tom Chiu, game level designer, a graduate student who already made several popular levels for Portal 2, Half Life and Gears of War.

Could you describe yourself and your background?

I'm a graduate student studying level design at the Guildhall at Southern Methodist University. I have a passion for video games, especially action games. I love developing games and will do this for the rest of my life. When I'm not working on games, I love traveling, cooking, and playing the guitar.

You've made several game levels in the past, what decided you to start designing levels?

I've always been fascinated by the way games immerse me into a different world. When I grew up, I realized someone must have created these games, and I began to pursue creating games myself. I started with modding Rogue Spear, trying out different disciplines like art and programming. It is only after I tried out different aspects of game development that I fell in love with level design. I love how closely I can shape a player's experience as a level designer. I crave the ability to make a player feel.

You've made levels for Source and UDK3, which engine do you prefer?

Honestly, both engines are top notch, and if you hear a level designer complaining about either, he hasn't worked with enough engines yet. I love the flexibility and WYSIWYG-style of the Unreal Engine and the workflow and the ease-of-use of the Source Engine. Both are extremely stable, scalable, and user-friendly.

How long does it typically take you to design a level, from start to finish?

I have the advantage of making levels for completed games, so the entire process is pretty streamlined. The time it takes depends on the quality of the game and the style of play. Typically for the levels I create, which are AAA quality and action or puzzle-based, it takes about 8-10 hours per minute of gameplay. There are some exceptions depending on the theme of the level. Vehicle levels take longer, and forest levels take a bit less time.

What takes the longest, designing it in the editor, or working out all the bugs from testing?

Depends on the ship date. In the deal world, I'd get lots of time to work out all the kinks. That really isn't feasible, so whitebox to alpha takes way longer than beta to rtm. In my experience, the level building takes longer than the bug fixing.

You've written a Master's Thesis about making a game with just 2 people, could you briefly describe it?

The actual thesis is still in the testing phase, so I won't divulge too much for the sake of the integrity of the project. However, I will say that we are following through an alternate game development method to the current one practiced at most game studios. We are seeing if it is possible to formalize a cheaper development process that complements the current system. Many successful cheap titles have already paved the way, and a formal process would make it easier for others to do the same. It's not project about competing against the current process. It's something that would benefit both the developers and the publishers. We are making an action driving game in Unity from scratch.

When you create a new level, how much of it is about the artistic process and how much is it about making the level be fun to play?

That often depends on the goals set forth by the project or lead. Making the level fun is extremely important, but honestly, aesthetics is also part of it. If a level doesn't reach the target graphics that the end-user expects, its aesthetics will detract from the gameplay. It really is hard to quantify these tasks into intelligible information. Basically, it depends on what you want to accomplish in each level. Not quite sure if I answered that one.

Your levels seem fairly popular, do you get a lot of feedback from users and how does that influence your work?

You can never get enough user feedback, so I try to throttle my best testers to ensure that I still have Kleenex testers near release. I take extreme care of the user's feedback when the level needs to communicate an idea to the player, such as when the player receives a new weapon, encounters a new enemy, the environment changes, or something of that nature. User feedback is the only way to ensure that the players are getting the experience you intend for them to have. Obviously I have instincts of how the gameplay is going to go, but it is difficult to account for everything that could happen. Sometimes emergent gameplay actually improves the overall experience, so I capitalize on that as well.

Is your goal to go work for a game design company or start your own studio?

I plan on entering the game industry as a level designer in the near future for a AAA studio. My immediate concern is to produce quality work for a company while learning from its veterans.

Are you working on any new project you can talk about?

I'm working on a medieval and steam-punk mix deathmatch level for Unreal Tournament 3 to demonstrate my detailing and multiplayer design skills for my portfolio. It takes place in a castle built over a waterfall, with water acting as a playing space instead of a hazard. That will be completed late August. After that, I will be writing a 3d game engine using the XNA framework to help me get a better understanding of game engines. That's all math and programming stuff, so I don't want to bore you with it.

Back to index

© 2007-2019 Patrick Lambert - All resources on this site are provided under the MIT License - You can contact me at: contact@dendory.net