Written by: Gabriel Tay – Producer/Lead Designer (Chrono Disfunglement) All content, 2013 © gabrieltay.com
Before proceeding, do check out this excellent article! It contains some informational tidbits on our development process!
This page is in the process of being developed to detail my experiences, challenges and the planning that went into our game, Chrono Disfunglement, which debuted the Single Target Rewind mechanic, in any game ever.
Jump to:
Concept
Level Design
Concept
This was in the Summer of 2012 when my team and I were discussing and brainstorming ideas of the kind of game we wanted to pursue next. I personally had an agenda to capture the zeitgeist of the gaming industry: the rise of innovative indie titles that focused on creating concise and truly unique experiences in an industry saturated with the standard fare of cookie-cutter FPSes, MOBAs, RTS and other action titles, was both an exciting and a refreshing outlook at the future of games.
Time seemed like a fascinating mechanic as we, in our daily lives, seem to always lack enough of it to accomplish what we wanted to do. Of course, it seemed like an uphill task both technically and design-wise as there were memory considerations to take care of (both on hardware and the player’s). In addition, we also wanted something that could let us stand out in the time fabric filled with great titles such as the critically acclaimed Prince of Persia, Braid and Blinx. If there was one thing in common among those titles, they all shared the ability to undo your mistakes by going back in time. We looked to the indie scene to see what else other games were doing with time. Most did not deviate from the aforementioned formula; some like The Misadventures of P.B. Winterbottom, took a different approach of using one’s past self to help his present self. But none seemed to use the flow of time in a more tangible manner.
What if we could make use of an event in the past to affect the present?
– Gabriel Tay, Lead Designer / Producer
I proposed an idea by asking a simple question – what if we could capture a moment in time and recalling it at will at any moment in time in the future? The idea leaned strongly towards an action-style game in which we had guns blazing and missiles darting across the playfield. By simply creating a physical ‘time bubble’, the player effectively took a snapshot in time (saving every action regardless of what it was). Actions would not be taken out of existence but continue as if nothing affected it. The point was then being able to recall this moment at anytime (which would recreate everything captured in the time bubble) at the player’s will. It seemed like we had a winning idea as it offered truly emergent moments that were entirely based off the player’s actions. We sat down as a team and discussed excitedly at the potential of such a mechanic; but it soon yielded design problems such as a player’s expected memory length to which they can remember that moment, determining what would make for such interesting scenarios, as well as physically determining a good way to deploy the time bubble. We put this aside and decided to approach it from another angle.
In a similar vein, we then thought of time flow in the context of a multiverse – what if we could split the timeline and physically enter an alternate time line (via a portal, because who doesn’t like portals?), change its future and merge these two timelines by carrying the outcome of the alternate timeline into the main timeline? It seemed to minimize the issue of physical importance and also offered an immediate outcome in which the player directly participated in. Again, it was admittedly a very intriguing concept that we thought was technically feasible; but once again, reality crept up on us about how we could design interesting gameplay and levels that was elegant and simple enough to understand, it proved to be a truly mind-bending task that we all agreed needed to be further simplified.
At this point, I thought I had a dug a hole too deep for the team to get out of as we were behind schedule without a solid game idea. The team was also beginning to wonder whether we were too ambitious, putting ourselves up to the grand and scary task of dealing in time in a non-contiguous way. We knew we had to find a solution without losing the vision we had so passionately aimed for.
The worst case scenario would be simply creating 3D Braid!
– Sean Chan, Technical Director
We knew we had to make decisions quickly and kept 3D Braid as our backup plan if all else failed. I was not very keen on losing grasp of the potential of the concept we had worked so hard to think through just yet. We went ahead and implemented our core time control technology based off Jonathan Blow’s excellent GDC talk of how he could keep so much data in time (this was going to be a challenge we had to deal with). The first time we could rewind/forward time at will in our engine, it amazed us even though we had seen it done so many times in games that dealt with time. Soon after, we developed our defining technology to handle a separate time stream that could operate seamlessly with the global rewind. Its first use was as a literal stream of time pellets which rewound objects that had been hit by it ie. if we moved a box from its original location, shooting the box would take it back to its original location without affecting anything else in the game world. It was exciting watching it work. Then we asked ourselves yet another question: “Does a player expect the global rewind to undo the rewind action on an object or does the player accept that in relative time, the object has already moved into the past?” We liked the idea of the latter actually as it made logical sense. But when it comes to game design, real world logic doesn’t always apply.
Do we undo the undone action or is the undone action already done?
Playing the game a few times with a simple scenario we developed using the aforementioned causality of time was frustrating. Even more frustrating was why something we knew that made complete logical sense lose its meaning when placed in the context of a game. We knew we had to tune the design to something players could readily accept and most importantly, expect. It didn’t take long for us to switch tracks and go for the gamer-centric approach which immediately made playing the game much easier and approachable.
I tasked myself and my fellow level designer to come up with more puzzle scenarios and basic levels to test out various ways we could harness the dual-time stream concept. The development was exciting as we soon churned out various scenarios that we personally hadn’t experienced before in any other game – experiences that ranged from slow-paced puzzle solving to more adrenaline-fueled dodge and chase games (we added basic AIs that could shoot and chase after the player). As much content as we had made, I soon felt that we were losing focus of what direction the game was heading for. Was it an action-platformer with players having to deal with external challenges beyond the puzzles? Was it a puzzle platformer in which players deliberated every move? Or did it lie somewhere in between that continuum?
We sat together as a team and took a good look at the current state of our game (which consisted of mostly neon-coloured cubes and spheres) and suggested idea after idea – each taking the game further and further away from its defining mechanic. We were ‘that’ close to losing it all and giving way to a Matrix-styled game in which you could stop bullets and do crazy stuff. We were losing the very soul of the idea and I was getting worried.
… To be continued soon.
Level Design
My personal highlight of Chrono Disfunglement was when I had designed a level which truly embodied the concept of the Single Target Rewind. As it was a mechanic so new and unique, even team members who’ve worked on the project were truly astonished by the eureka moment the level design presented. One of my team mates even remarked that he was about to give me a good scolding for putting up an unsolvable puzzle (which he did solve before getting to me). Playtesters who had the opportunity at a playthrough of our game frequently mentioned this particular level as a standout due to its ingenuity and elegant solution which really showcased Single Target Rewind ability.
In case you skipped reading what the Single Target Rewind is: it’s the ability to rewind a specific object in both space and time. If two objects, co-exist in the same space in time, they will mutually destruct, destroying both objects, thus erasing them from time. Mutually destructed objects send out a non-lethal shockwave that physically affects objects caught in its blast radius by launching objects as well as the player into the air away from the blast origin.
Playtest Reactions
Here are some comments I received from playtesters:
“… finally figuring out how to beat it was the best moment.”
“What was my favourite thing about the game? The unique puzzle at the end, more of those please!”
“The last puzzle showed what really could be done with the game”
“The final puzzle, because it was the first Digipen puzzle game to SERIOUSLY stump me, and I felt like an idiot for not realizing the answer once I figured it out. Perfect for a puzzle game.”
NOTE: Screenshots are taken from the earliest incarnation of the puzzle which is presented in the most unadulterated way.
Objective and Planning
The level was crafted as a culmination of all the skills and techniques learnt using the Single Target Rewind ability which included sections requiring a combination of either one or more of those skills to overcome the puzzles. This level was probably the most ambitious in design as it expanded the playable area to about 3 to 4 times as large as the earlier levels. Such a scope would easily confuse players in terms of navigation in a 3D environment and I took extra care to plan out the necessary props and key moments to guide the player. I designed a big set piece moment, involving a sliding section to set the tone of how the level would be: a long descent that gave the player a glimpse of the end goal before sliding away and out of reach of it. Using a combination of lighting and motion, the level was visually divided into clear areas that they know they have traversed so as to reduce the likelihood of redundant backtracking. In addition, I made it a point to drive home the idea of being time-aware ie. that every action in the past can potentially be a solution in the present.
The design planning process was almost entirely done on paper simply because of the ease of quickly drawing out layouts and marking the sections accordingly. The idea was to have a clear representation of an annotated map that highlighted important points and the sub goals to accomplish before the player can proceed. Landmarks and key props for player guidance were also marked out to establish the ideal or expected route a player would take when playing the level. In addition, a level flow chart was put in place to not only determine level progression, but also understand the player’s mental state and threshold for giving up or moving forward. For example, the player overcomes a particularly tricky puzzle, do we want him/her to jump straight into yet another challenge or is do we deliberately lay down a seemingly meaningless stretch of ground to both physically and mentally mark an accomplishment as a well-deserved breather? This helps me identify potential problem areas and iterate on them accordingly.
As previously mentioned, the last section involves a 3 step puzzle with limited resources and also serving as the final obstacle to the end goal. This particular puzzle was as much a challenge to design as it was for players to play it as I literally had to play out the scenario in my head in a way that made sense and also not be open to loopholes.
The puzzle design was originally done on pen and paper; laying out the elements I would need for it to work and quickly sketching out possible scenarios that may arise. With that done, it was only a matter of actually implementing the puzzle in game and playing through it many times and in as many possible ways I could think of. It was critical that I kept a mindset of a new player who has already experienced a good chunk of the game mechanics up to this point in order to not approach designing it with developer blindness.
The puzzle went through about 10 iterations in game as each play through seemed to expose a loophole which completely broke the puzzle – problems such as providing too many resources, incorrect platform placement, unexpected outcomes, convoluted solution paths – all these were challenges I faced before I settled on the final design which can now be found in the game. It was interesting to note that slight adjustments in positions can affect the perception of a puzzle, even though no elements of the puzzle were added or removed.
Implementation, Testing and Understanding
Without spoiling the solution, here’s a brief overview of the parts of the puzzle that players would need to take note of.
The first button, when triggered, releases the forcefield blocking the portal.
The second button, when triggered, releases a drawbridge that links the lower area to the isolated platform housing the portal.
Accustomed to the fact that boxes are primarily used to hold down buttons, the player assumes he/she would have sufficient resources to solve this puzzle easily. However, the player will soon realise that there is a chasm separating the upper area to the lower area. Most players attempted to push the boxes only to see it fall straight down regardless. Some explosive force was bound to be required.
Before proceeding, I wanted to understand the thought process and psychology of the player up until this point of time having learnt and picked up the necessary skills to solve the puzzle. It was imperative that players adjusted their paradigm; simply put: to see beyond the third dimension. I spent time researching the average memory capacity and duration of a typical person to judge how I could best present a puzzle which cut out redundancy without losing the essence of the objective. I wasn’t overly worried about players giving up quickly since our inertia for retrying was almost non-existent with the Global Rewind ability. This encouraged players to try out different ways to solve the puzzle without having to click on buttons or enter any menus to restart – they simply rewound to the point where they best decided would be ideal for reattempting the puzzle.
Most players at this point would attempt to use mutual destruction to launch the box to the lower area. The preconceived notion that they would require 2 boxes to solve the puzzle and having mutually destructed a pair would only leave one (which would be the one they launched) often was a point where they started to doubt themselves. This point was the pivotal moment in understanding that the box can be in a past position while in the present – the very idea of seeing beyond the third dimension and recalling the events of the past. Once players realise and overcome what they think is physically impossible, the solution becomes crystal clear.
