Greetings! My name is Raymond Ng, Technical Designer/Art Manager on The Channeler and today, I’ll be discussing the design process that the team went through from January to July. As you’ll see, video games undergo tons of iteration and design, and this game is no different. The final product you see on our site is very different than what was initially pitched.

The initial design

The Channeler as it is today is extremely different than what it was initially pitched back in December of 2015. Originally, the game was all about using VR and eye-tracking to create an experience where the player starred as an agent of a special organization dedicated to resolving problems between the living and the dead with the help of other spirits from reality and myth.

However, we quickly discovered that VR headsets with eye-tracking didn’t exist and the cost of having a custom VR headset with eye-tracking built-in was estimated to be over $10,000 for a single unit. On a team of possibly 10+ students in a graduate school. This led to a choice between VR or eye-tracking and so, eye tracking was chosen as the big focus of our game’s experimental mechanics.

Preparation for vertical slice

During the pre-production phase, the entire team met up to discuss design ideas for mechanics and locations for environment art inspiration. In terms of premise, the first pitch was a seedy city that was almost slum-like set in east Asia. The idea that caught our attention was a game about ghosts set in a haunted version of Kowloon’s infamous Walled City. After all, this game was about ghosts and using your eyes in unique ways to play the game. We immediately set forth building and compiling all sorts of data and references ranging from myths and folk tales for ghost companion ideas, to real world locales for environment art inspiration.

The game was presented to the faculty as having various mechanics and features for vertical slice:

  1. Using your eyes to guide spirit companions around to do things with other NPCs or interact with the environment.
  2. Using your eyes in social situations by looking at things to create unique reactions or effects. Example: looking at a ghost’s tattoos will cause the ghost to make a remark about them.
  3. Ghost companions have their own unique personalities and ambitions. Talking to them would initiate Mass Effect style dialogue options. There would be a dialogue wheel that changes the tone of your dialogue options.
  4. Combat involved using your eyes to direct companions around and get them to perform unique attacks in a puzzle style manner. The player at no time should be the one attacking, only slowing down or snaring the enemy.

(All this in a game that would take effectively 7 months to make!)

The biggest thing that came up during this phase was two original characters: a Chinese hopping vampire that was part of a gang of Chinese vampires running the city, and a teenage ifrit, a djinn of fire and smoke. The purpose of these two characters were to be companions that you could talk to and give hints when exploring the city. They would also be voice acted because we were aiming to have the Mass Effect style of companions.

(Early concepts for the now scrapped companion characters)

So after coming up with our list of inspirations, designing the characters and coming up with exploration mechanics, companion abilities, dialogue, and boss battle mechanics, we set out to build our prototype in preparation for vertical slice.

Our game made it. Sort of.

So we presented our game for vertical slice. If the game showed potential, our game would get green-lit by the faculty and we would continue into July. If we didn’t make it, the game would get cut and the team members would be scattered among the remaining teams. Our game made it. Sort of. Our game, which we pitched as attempting to innovate, was deemed too iterative by the faculty. They would green light our game with a major caveat: redo everything from the ground up. That meant throwing out everything, character dialogue that was already written, designs, level prototypes, voice acting, you name it.


(A concept image of what was imagined for the main level. This would change in May/June.)

Rapid prototyping

After vertical slice, we basically redirected our ship. We turned off our story and dialogue engines, and diverted all energy and focus attempting to innovate with experimental mechanics. During this phase, we began rapidly prototyping mechanics/puzzles. Each week, we would design up to three mechanics with a puzzle example, create a gif that lasted five-ten seconds, and present them before the team. As we presented them, we would explain all the things that would happen. Afterwards, we would begin to prototype the most promising mechanics with no art input-that would be done later.


(Weeping Statue Puzzle. If a statue is on the screen and the player blinks, the statue will move towards the player. The goal is to move statues onto pressure plates to progress. This mechanic would later be used in the ante-penultimate puzzle.)


(One of our prototypes, the floating keys, required using your peripheral vision to recognize shape and color. The goal is to find the matching key hidden within a group of dummies that suddenly appear. If the player looks at the wrong key, all the keys will disappear and the player must start over the current puzzle stage .)

After we built our initial repository of prototypes, we began to play test them internally and publicly. Based on the positive or negative feedback, we decided which prototype to keep and which to drop. Afterwards, we brought the well received prototypes into the next stage: programmer and designer refinement.


Refinement is a big deal. We decide where a puzzle will be presented, how to tutorialize it, how many iterations the players will play, how to frame it and the premise around its existence, the art assets needed for it, animations (if any), lighting considerations, which programmer will work on it with which tech designer, and bug fixing. TONS of bug fixing. After refining, we began to playtest the prototypes even more intensely and pushed to bring it to more internal play test events with the school.

By the time mid-May rolled around, we had a handful of polished and well received prototypes. Now that the mechanics and puzzles were in place, it was time to put them in the level and frame it. We ultimately decided to go light on the story and focus heavily on the gameplay as The Channeler was all about using an eye tracking device to play games differently.


(The floating key puzzle re-skinned to be haunted bottles in our final build. This screenshot was taken in editor. The functionality remains the same, but the theme has changed since prototyping. The glowing circle is the eye reticule that appears when the player looks at an object that can be interacted with using their eyes.)

Quantum leap forward

With all of our puzzle mechanics in place, we decided it was time to build out and prototype a level needed to guide the player and introduce the player to our game. Our previous attempt at creating a sprawling city style level would have been way out of scope so we had to reduce it tremendously. I worked with the art team to block out a prototype level that would later be used as a basis for our level in the release build. Later on, I would refine the level by adding more paths and rapidly adding various proxy art assets that the artists had made.

This is where the need for proper art assets began to show during one of our presentations. Later, the artists gathered together and held what we called an “art jam” where all the artists would pick one day out of their school schedule, and bring art assets through the development pipeline. Since all the artists were working together in tandem, this allowed for easier planning and removal of dependencies (which is especially nice if you’re an art manager).

(Left: The level diorama that was built to visualize our art. I worked with the art team to build this quickly to give our artists a stronger sense of what we hope to achieve.)

(Right: The final image of the main market section of where our game would take place. After tons of polish, adding blobby ghosts, crowd AI, multiple lighting passes, and high quality art assets, it finally looks like something out of a video game.)

With the implementation of art jams, our next presentation was deemed a quantum leap forward by the faculty and we would begin to hit everything in stride. In fact, progress between each update seemed to start skyrocketing compared to the previous months. Now that the team could see the potential of the project, everyone got really inspired and want to put in extra hard work!

More playtests

With our game’s puzzle mechanics complete or near complete, it was time to begin playtesting. We held playtests at a bar called The Geekeasy, UCF’s main campus, and at Iron Galaxy‘s Orlando office. Iterating on feedback and making bug fixes between each of them. During that time, we took all forms of feedback and converted them into actionable tasks. It was also during this time that we began to put major focus onto sound as I started mixing sounds and recording VO with various classmates for The Channeler.


(Playtesters at the Geekeasy. It was our second playtest there and we received tons of useful feedback!)

With more tasks now leaning on the tech designer side, task hours for the tech designers would skyrocket. The occasional crunch day would become inevitable if we were to complete every single task and have the build debugged and ready for another playtest.


After we finished our IGO playtest, we were effectively in a freeze. Small tweaks to dialogue and minor sound balancing were left, but new art assets, content, and features were no longer being worked on as the programmers began looking into any remaining bugs that could be fixed in time.


This brings us to today! We’re less than a week away from our final presentation so I thought I’d have this blog post written of our journey–from the design perspective!


(Our game poster!)


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s