Post Mortem

The Inside Man is a 3D stealth/strategy game in which players plan heists ahead and execute them to steal treasurous things. Players need to place traps and distractions to those AI Yarger Bots to make their way in and out and perform the actions with right timing.

Our team was inspired by Jacob Zion’s project 2 game Stealth Golf. The initial idea was to make a 2D top-down stealth game where players have some time to manipulate the map before actual gameplay. We thought that it would be fun if combining a stealth gameplay with a strategy one. As for player impact, we hoped that by giving players more information and letting them manipulate guards before acting, they could gain the feeling of satisfaction after successfully executing their long-sequential plan. Also, misoperation or unexpected events during the action phase would add to the drama of the game.

In the first week, we were not clear about what exact elements we need in the game for players to manipulate. The only one we thought of was the frozen trap which can stun the guard stepping onto it for 3 seconds. By setting the frozen trap in the way of guards in advance, players can create a phase difference between the guards’ patrolling patterns, and use that to get through them. To showcase the stealth core gameplay with the frozen traps, the first prototype we created was in 2D top-down perspective.

In the following week (milestone 1), Zhenyuan Zhang came up with the idea of making the playing phase first-person perspective to give players a more immersive experience. Also, we found it super cool to have the traps go off at the same time in the playing phase as when they put them in planning. Together with that, players can preview all guards’ movements before doing the actual stealth and form dedicated plans of themselves. This turned out to be the core mechanism of the final game.

Instructors and peers gave us lots of useful feedback for our milestone 1. The most valuable one was that we didn’t convey the “scheduling” idea clear enough. We were suggested to implement a timeline for memorization so that players could get the idea at the first glance. It took us about a week to implement the timeline visualization and it worked like a charm. In that week we also playtested our level transitions a lot, which solved a bunch of bugs regarding pause locks. At this point, we had our core gameplay mechanics in a more functional state, and as such we decided to begin working on our first real heist beyond the tutorial. We decided to make the heist take place at a shipping warehouse and have it structured so that the player first had to make their way inside the warehouse and then into an office to steal some documents. The first prototype of the warehouse heist was small but effective, and the environment allowed the player to navigate through the level in a few different ways. 

Though the timeline and the planning system were well-implemented in milestone 2, the game still lacked mood, theming and juiciness. The most important issue we thought of was the lack of guidance in the tutorial. We kinda taught players through texts, but few playtesters took them seriously. Also, the tutorial was broken into four parts, which degenerated players’ feeling about being a spy. So we redesigned the whole tutorial and made it into an unified heist called “Prison Break”. We believed that making the tutorial into the story would make players more engaged. 

After completing the alpha build, we gave it to our roommates and watched them play without giving any instructions. We got several feedback, including complaints about motion sickness and the guards being too easy. The instructors’ critics were still about juiciness and theming because there were still no improvements on sounds and the design of the warehouse heist. We started by adding in background music as well as sound effects to make the game feel more alive. We also decided that the warehouse heist needed a complete overhaul in order to make it more interesting and fix some design issues from the initial prototype. The new iteration of the warehouse heist had the same general structure, but the scale of the level was increased dramatically and more complex environments were added in each stage. We had a lot of interesting ideas for ways the player could enter and escape the warehouse, and eventually we settled on the idea of being carried in crates to enter and escaping via delivery truck. We also began designing a second heist which would have been a bank heist, but we were unable to overcome certain challenges in the level design before the deadline.

Based on the feedback from our beta submission, we knew that we had a good amount of content in the game already, and instead we needed to focus on polishing the game and improving the theming and juicing. As part of this, we also decided to cut the bank heist from the final submission of the game, as the issues we were facing with the level design would have resulted in a product that was very buggy and overall lacked polish. We made many quality of life changes, including allowing the player to replay their plan without having to plan it out each time. We also did a lot of work on balancing the game as we received a lot of feedback that the warehouse heist had become very challenging. The dialogue for both the warehouse heist and the prison break heist were also edited to guide the player better and more appropriately fit the theming of our game. We added more realistic textures and lighting to all stages of the game to give it some extra polish, and added even more music and sound effects. By the end of the deadline for our gold submission, we had polished the game and gave it a sense of authenticity that we could be proud of.

During our group meetings we used our time very efficiently to determine what exactly we wanted to accomplish for each iteration. We also had several productive discussions when deciding how we wanted certain mechanics of our game to work. These meetings and discussions allowed us to communicate each of our individual thoughts and ideas in order to create a clear shared vision for the game as a team. This also allowed us to have very effective iterations as we were able to address all of the feedback we received as a group and determine which changes to prioritize during development.

We did have some issues with deciding on the overall scope of our game. We spent a lot of time developing a second major heist to add to the game, but after running into several issues with the layouts of the levels and mechanics of certain areas we decided that we did not have enough time to polish it in time for the final submission. This resulted in a lot of lost time and work that did not end up being represented in our final product. 

In order to ensure a better result in future creative endeavours, we have learned the importance of being more realistic with our ambitions and focusing on improving and polishing core features instead of trying to add on more.