Why Project Stealth failed but Antarctic Ranger is succeeding

Preview for Project Stealth

For those of you who aren’t familiar with my game design and development projects, prior to the Christmas period in 2019, I was working on a game that was codenamed Project Stealth, intended to be a stealth game which intended to demonstrate my level design skills but I felt as if I was struggling to pull off such a game. I decided to begin re-educating myself in game programming to better improve my ability to implement the game mechanics that I have designed, especially with how sloppy the visual scripting for Project Stealth was. Since starting to learn programming and visual scripting skills, I’ve stopped working on Project Stealth and began a new project dubbed Antarctic Ranger, which I’ve found myself to be much more proud of and more motivated to continue developing it.

This blog post intends to analyse exactly what mistakes were made with Project Stealth, which was designed and developed with considerably less knowledge than I have now, and how the lessons learned from Project Stealth in addition to quickly brushing up on my programming skills over the course of last month (December 2019), have been applied to Antarctic Ranger and how much more successful the newer project was as a result of this as well as other things that were learned outside of both Project Stealth and Antarctic Ranger.

Early Gameplay Prototype for Antarctic Ranger

What happened while developing Project Stealth?

I went into making Project Stealth freshly after completing my Master’s Degree with the hope that I could make a very good level where the player would use the level design to successfully sneak past the AI, which was initially made by following a tutorial on AI programming in Unreal Engine 4 despite having minimal knowledge of game programming fundamentals at the time. This was coming from little to no experience or knowledge in level design despite specialising in Design at University.

Adding to this, I had very minimal understanding of what I needed to be doing for a Game Design Portfolio and most of the research I did led to me finding portfolios containing games that were completed, including having well-done artwork and were very polished. This made me assume that I needed to have completed, polished games that were not only functional, but were also great graphics-wise. As someone who mostly engaged in Design, dabbled in programming and did little to no art while I was a student, this intimidated me as I felt as if I needed three to four completed, AAA-standard games.

This led to using free-to-use art assets and using them to create buildings, obstacles and so-on without greyboxxing the level first, thinking that the game had to look polished straight away. Furthermore, my programming skills were very limited, meaning the gameplay turned out very sloppy, leading to having difficulty with implementing the game as envisioned.

Project Stealth Gameplay Video containing pre-made Character and Environment Assets

After creating this version of Project Stealth, I demonstrated it to a handful of game and level designers who were attending EGX in October 2019 and was advised that in a Design Portfolio, I was expected to demonstrate the process I went through to design and develop my games rather than simply showing off amazing looking, completed games. After hearing that (and enjoying the day out at EGX), I decided that the original level that I was working on wasn’t working out well and decided to start the level design again from scratch. The evidence of paper-drafts and new level blockouts can be found here http://sigrothian.co.uk/project-stealth/

After running into difficulty with the gameplay programming, that being too reliant on tutorials and leaving my code very messy in the process, it just wasn’t working out anymore and was really taking a toll on my confidence in my abilities as a game designer and reminding me of the negative experiences I had at University with programming, the harsh criticism in my abilities from other students who I clashed with and even a lecturer questioning how I managed to pass a mandatory programming unit when I was in my first year of University.

The overpopulated Event Graph for Project Stealth’s Player Character. Just too much going on.

What I wish I knew when I was working on Project Stealth?

Reflecting on Project Stealth’s failure, re-educating myself in game programming (a journey which started here https://www.linkedin.com/posts/smsmith195_today-i-came-to-realise-just-how-important-activity-6603794039752007680-OG2p ) and the lessons I have been learning recently with Antarctic Ranger (which we’ll come to later), I wish I’d had better insight into what is actually expected in work presented on a Game Design Portfolio, which would have been much more beneficial for building Project Stealth as it would have at least assured me that I didn’t need to worry about how good my games should be regarding the graphics and allowing me to focus on designing and implementing gameplay and level blockouts rather than how good it looks.

How has Antarctic Ranger’s design and development differed from Project Stealth?

While Antarctic Ranger’s development started while I was brushing up on my programming skills, I made a mistake very early on where I attempted to implement a fully animated character to represent the player but was faced with a lot of complications, mostly based on the free-to-use character assets that I had access to not being designed for first-person games, which resulted in me deciding to simplify the player character by using the built-in assets used in Unreal Engine 4’s First-Person template and a free-to-use gun package without using any animations. While this was yet again, worrying about the graphics, rather than just going for it, I questioned if I really needed to worry about the graphics when making prototypes. The answers I got were very assuring, one advising me the focus on making sure the gameplay worked with cubes and another stating that making complete games is good but it didn’t matter how good they looked graphically.

Following the advice received when I asked the question, I continued to develop Antarctic Ranger, prioritising gameplay over how it looks. The only aesthetics I needed to worry about were the information being shown on the player’s heads-up-display and changing the colours of enemy character “cube bodies” to communicate when the enemy had been defeated.

From a programming point of view, I’d learned about encapsulating code into functions which helped to simplify the game’s programming and making scripting the gameplay more accessible than it was when I was working on Project Stealth. I got used to working that way to the point where I’m now starting to be able to script my games without relying on tutorials for everything, only needing to do research if I became truly stuck and looking for alternatives within my skill level if I deemed my original goal to be too difficult.

Part of a Script using Encapsulated Functions in Antarctic Ranger’s Player Character Object

While Antarctic Ranger is far from over, there are still plenty of more lessons in Game Design, Level Design and Game Programming that I’ve yet to learn which will be reflected upon constantly through my Twitter and LinkedIN and soon, I hope to bring a full breakdown video going over the design and development process, demonstrating my thinking and solving Design challenges when Antarctic Ranger has been completed. For now, it’s safe to say that the development of Antarctic Ranger has gone a lot more smoothly than Project Stealth so far but there are still challenges ahead and I’m ready to take them on as and when they arise.