“Unto The Horizon” Postmortem – Week 2

Hello! You’re at the week 2 postmortem for “Unto the Horizon”, an ETC BVW Round 4 project at Carnegie Mellon University. Here’s a quick glance of what we did for week 1:

  • Round 4 is a story-telling round.
  • We’re using the Cave Autonomous Virtual Environment (CAVE) room. It is a room with 3 big screens and a moving floor.
  • We’re telling a story about launching a rocket. We aim to invoke emotions of accomplishment and the joy that comes with it.
  • It’s an asymmetrical experience: One team will be in the CAVE as pilots, and the other team in another space as Mission Control. Plays like Keep Talking and Nobody Explodes.
  • Feedback: Lack of story, no interest curve, not intense enough.

You’re always welcome to go back to the previous blog post to know more about this project! Without further or due, let’s begin our week 2 development!

Going into Deeper Space Or Digging a Deeper Grave

We kind of already knew that it was missing something even before we received feedback, but we just didn’t know what the missing part was. We looked into the feedback and one in particular hit the spot: not intense enough. It could be that our game had no actual risk of failure, or failure, the case in which the rockets explode and astronauts die, meant little to noting to the guests.

Failure meant much more in Interstellar, The Martian and Armageddon. In Interstellar and Armageddon, it meant the death of humanity; in The Martian, it could’ve wiped out NASA’s (remaining) reputation, plus Wattney’s team would feel forever guilty for they’ve left him to die on an alien planet.

So, we decided to make our own “Interstellar” as well. Although undeniably it’s a bit cliche, but given the time we have left (1 week), cheap and effective is probably the best solution.

It took us only one day to come up with a back story that didn’t require us to make big changes to our main game:

In the near future, a fungal parasite spread across Earth, eating up all vegetation on the planet. Scientists tried to develop pesticides, but it didn’t help. At the same time, humans have found a new planet that could nurture life; they’ve also found out a way to code human genes into plants, a way to build an ark for all life on Earth. Time has run out for life on Earth, but not for the possibility on another planet. You, as the team in charge of this final space mission, will ensure that life gets to move on in a new world.

This is a story about saving whats left of humanity, even with knowing that in the end you will sacrifice. It is a story about accepting the harsh truth and moving on.

Thinking back on this, I think all of us on the team would probably agree that we were either doing something really epic or we were just digging a deeper grave for ourselves, and I’d say that many times it felt more like the latter. That being said, I am still surprised that we were all more than satisfied with this almost-suicidal plan and managed to pull it off.

Having a clear picture of our game really helped me a lot with both the music and the paperwork. I quickly realized that this game definitely needed an introduction / epilogue video to get the guests quicker into the situation. Taking in the story and our budget and time into consideration, I rapidly drafted out a storyboard and asked Muhammad to come up with still visuals.

The storyboard for our epilogue (1)
A comparison between storyboard draft and results. Muhammad did the still and I animated it in After Effects.


For music I immediately went back to Interstellar, but for technical aspects rather than the melody itself. I wanted to learn and understand how Hans made his music sound majestic and epic, so I broke down his music into parts (“Mountains” in particular, not “No Time For Caution”). Here’s what I’ve identified:

  • Reverb & Echo: Using reverb adds space into the music, and echo is a hint of an extremely large space. The use of both will help creators define a space that resembles a large church, which is large but not too large. Also the strength of reverbs and echoes helped determine the materials of the surfaces of said space. For our game is also about space, I wanted to be gloomy and spacious so I tried to recreate a space that’s even larger than what Hans had with Interstellar.
  • Panning: Panning is essentially offsetting an instrument / source of sound by a certain angle horizontally to the listener. Panning also helps creating a spacious feeling to the music as the piece doesn’t sound like coming from a mono-speaker but rather around you. Doing this is not so hard as it’s a built-in feature in Logic Pro X; the only thing that I needed to be careful is to balance the experience so that one channel wouldn’t overwhelm the other.
  • Bass (Bwaas): The intense bass performance in Interstellar deepens the experience and perfectly reflects the weight of the story. It helps the narrative more than the music itself. Despite the weight of our story being similar to that of Interstellar, we didn’t want our guests to feel anxious or be in a “it’s do or die” mindset; we wanted them to think that “it’s worth it for humanity, and I’m a hero.” So I kept away from that and only added in some bass (mostly from percussion) in the end of my piece.
Image from iOS.jpg
Me and the team discussing details in the sound room

Just like the game itself, I went through countless iterations with the team. I started out visualizing in my head what stars would look like in space, and how beautiful it could be. I immediately thought of a meteor shower: Visually it’s stunning, and since people wish upon shooting stars, a meteor shower really bears that concept and perfectly reflects the emotions of our game. This is where the downward [F# E D A] came from.

To incorporate the whole “rocket” theme, I then came up with a verse that goes constantly upwards, resembling the rocket leaving the Earth.

The whole game ends by snapping into pitch black immediately when the music stops. Then it brings out an epilogue told through a conversation between a mother and a daughter. We wanted the guests to imagine themselves as the ark itself – traveling alone in the wide, dark space, carrying hope to a new planet. I wanted them to understand that even after the darkest hours, morning comes again. For this emotion I decided to cut out all instruments, leaving only a piano to whisper in the vast space.

You can listen to my music here!

Puzzles and Interactions

I kinda knew it at an early stages, but I failed to let everyone know at the early stages of my problem. The problem was that I simply didn’t have enough time to come up with more than two sets of puzzles. I realized that it was hard for me to design puzzles that are both easy to understand and somehow-challenging to solve. And I was merely just following what they had for Keep Talking. At the end of week 1 I only had two sets completed and half-tested.

Since we overhauled the story, I had to put more of my attention on rewriting the story and scripts. Although I did still iterated through the puzzle sets I had, I didn’t invent more. The majority of these changes were focused on the user’s reading experience, like revising vocabulary for precision of language, changing font, and decorating the document so it would look like a real manual.

For interactions, we basically built off from what we had last week, and finalized the names of the buttons. One thing we kept in mind was that we didn’t just want the interactions to be realistic; we also wanted them to make sense with the context. We used a few slide-bar and knob phidgets for pre-flight adjustments, and we faked buttons by carving holes in a cardboard box then stashing a keyboard in it. The player will then press these buttons for puzzles presented later in the game.

The console we built out of a keyboard, some phidgets and some cardboard. We didn’t have time to polish the appearance but it serves it’s purpose well.

Something worth noting is that I stacked in a lot of fake pages at the intention of making the document look and feel scientific and authentic. These pages had nothing but the “lorem ipsum” gibberish you’d find somewhere in a blog creation webpage. While many playtesters loved the idea for making the experience even more realistic, there were also many guests think that it was meaningless and slowed down their speed of understanding the document.

If I had a second chance to go back in time and do it again, I’d probably still do the same. I don’t believe that these bogey pages could effect their reading abilities that much. But of course, I would also make adjustments, such as greying-out these useless text, I’d also add in a few pictures and diagrams to enhance the authenticity. I believe this pair of solutions would increase both readability and detail without hurting each other.

We tested out the console with the manual, and we found out that people were still overwhelmed by all the instructions and labels. We first took a mild approach in which we simplified a numpad-input sequence into just entering zeroes and ones. This adjustment alleviated the problem in terms of difficulty, but the current design requires a 100% correct answer to proceed, which turned out to be too unforgiving for guests. After discussion (and acknowledging time pressure), we determined that the story should prevail any other aspect of our experience. In the end, we removed all possible error catches in our game; that is, all input will be accepted, and succeeding events will push on.

To even solve this problem further, we decided to put two actors (NPCs) in our experience, one in Mission Control as Flight Director and the other in the spaceship as co-pilot. This allows us to control the flow of the experience, and give extra help should guests be trapped in puzzles.

Despite being critically short on number of puzzles, we were shocked to have learned that two puzzles could easily take up to 10 minutes for people to solve, and we’d only have 8 minutes at max for presentation. Knowing that we were actually at a good position in terms of puzzles, I could finally move my attention to what was perhaps the most important element of our game – scripts.


Before week 1 seminar, we had a simple script that was merely just directions from Mission Control. While in reality a conversation between MC an the pilots would sound dull and lifeless, which is what we had for week 1, it just wouldn’t make a good story to tell.

Image from iOS (1).jpg
Rehearsing in the CAVE room while waiting another team to come off. Muhammad is apparently tired.

I believed that to make a dialogue sound lively, one of the most efficient ways is to include exclamations and interjections such as “umm”, “huh”. While these interjections could help make the dialogue sound better, it is really up to the actors who needed to perform it out. To keep our team members on the same page, I made some remarks and notes on the playbook so that they could know what emotions they should be looking for.


It would seem like we came a long way from where we started. While I cannot agree more, I also believe we still had a long way to go.

Our final guests (the ones in the demo video) were sometimes confused about what to be done at moments. Luckily, with the help of actors, they were able to untangle themselves quickly and solve the problem.

It’s funny that the guests didn’t know what they did (shoot the ark and sacrificed the pilots) at the end until they found out seconds later. The Mission Control team was furious with awe and sadness as they realized they killed their now-frozen-space pals; but other than that, they found the experience compelling, unique and fun to play.

Finally, to review the goals we set out for:

  • An asymmetric CAVE experience: Check. We made a game that requires two parties in two different settings to play; what we didn’t do well was the balance of significance between the two rooms. In the end it felt like Mission Control was a non-essential part of the experience.
  • A Mission Control that involve the entire class: Cross. With only two puzzles we managed to confuse our guests for long enough. We didn’t feel like we were in a good position to move forward on this objective.
  • A compelling story experience: (???). Although we were pleased to see that guests felt sorry for their own decisions, that wasn’t the main emotion we were after. We wanted to tell a story about people having the courage to give up their friends’ and their own lives. While self-guilt was an expected feeling during the process, “determination” was the final answer.

On a side note, we never took time to figure out the voice connections until demo day. In the end we had to add everyone into a WhatsApp chat group and voice call everyone. We also ran into some microphone feedback issues in the CAVE room so we needed members in Mission Control to wear headphones (quite realistic though!).

That concludes the postmortem of “Unto The Horizon”! If you have reached this part, you have my greatest gratitude. At here, I’d like to thank Muhammad for his beautiful artwork, Min for his mad modelling skills and Shawn for making all of this happen through his outstanding programming knowledge. I’d also like to thank Sherry and Jue for lending their voices for the intro/epilogue cinematics.

“These people, just like you and me…
They gave up everything they had,
For everything they’ve ever loved,
For everything they’ve ever cared about,
For you, and me.”

From left to right: Muhammad, Shawn, Min, and me.

Published by

Brian Teng

Game Designer | San Jose, CA

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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