Gregor Smith - Game Designer

The Last Dragon

The last Dragon is a 3D, third person flying game where the player controls a new born dragon n a child’s storybook as they fight against the evil corruption that has taken their family.

I was the Game Director & Lead Designer on this project along side 5 others. We built it in Unity and put particular emphasis on making the flying mechanics engaging.

My main work was on the flying controller, camera, player experience, and level design. Additionally, as Game Director I led the vision of the project and managed meetings with the team.

The project started with an idea inspired by the movie How to Train Your Dragon and games like The Last Guardian. In particular I wanted to make a game that evoked the same feeling as the ‘test drive’ scene in HtTYD. Upon sharing this with those who would become my team members we decided to pursue the project as a UX and narrative focused game. We would dedicate ourselves to making the flight extremely enjoyable and have the player discover the history of a lost civilization as they fly around.


Some of the major challenges on the game included our lack of an art team. Early in the project I did the majority of the art and animation, particularly for the dragon. However, later in the project we overcame this hurdle by changing the theme of our game. We decided to set the game as a story being told by a parent to their child, with the things in the story being very simple toy-like objects. This matched our art capabilities perfectly! By December 2018 we wrapped up the project with the new art style.

A second design challenge we faced was the control scheme for the dragon. We drew a lot of inspiration from other flight-focused games like Ace Combat and the early PS3 game Lair. However, the latter utilized motion controls that we were not going to be emulating and the former was a game about flying a machine, not a creature. This ‘Creature-Not-Machine’ issue has become a key part of our design for the control scheme. We want the controls to feel more like the player is flying a creature and not a machine. The key things to consider here:

  • Meter-based speed. The dragon’s speed needs to feel more variable with input than like a simple accelerate, decelerate system.

  • Interaction with the environment. The dragon interacts with their surroundings, like their wing clipping the ground and sending dirt spraying.

  • Input-based animation. The dragon’s flapping frequency was something we tweaked for a long time. We’ve found that flapping needs to be tied directly with the player’s input. The same is true for most of the dragon’s animation. They enter a braking animation not only when the player is slowing down with a button press but whenever the dragon is slowing down itself (like when approaching a wall).

Level Design

In the final 3 months of development I took on a role as the primary level designer for the project. This happened as our development team had spent much of the past year working on the same parts of the game and wanted to try doing different things to avoid burnout. As I had only limited level design experience and we were redoing many of our levels to better fit our new narrative and scope, this meant I had to learn very quickly.

I started to build the levels open-ended with many paths to the final objective but this became complex. Players had difficulty keeping track of where they were going, often turning back in the direction they had just came. Furthermore, the amount of time we had available did not enable us to create particularly levels with multiple paths. After a few passes on the levels I went back and rebuilt much of them to be more linear, allowing me to spend time polishing the space. Fixing the issue of players getting lost took a little more design work. I took a three-fold approach to the problem:

Path: Each level has a path build of some type of rock or hoops that guides them down the correct path.

Lights: The levels have lit objects that go out when the player flies passed them. Not only does this act as a bit of a reward for progress but it means that the path the player is going is always lit and the path the player came from is always dark.

Color: To help the player track where they are in the level, the areas have distinct color schemes. Again, this allows players to look back and identify where they were opposed to where they’re going.

Overall this solved most of our guidance problems.