Back in June 2021, second year of Game Design had just wrapped up. A faint idea of what would become my Year 3 FMP started to formulate. Named Project Worlds originally, the game went through a heavy ideation process throughout the Summer of 2021. I really wanted to break the mould with this project and take the opportunity to create a game that was different from anything else out there.
The ideation process took me to many places, this game was originally going to be a VR only experience where you would fly to an island and be tasked with building up the island with buildings and a whole eco-system. Another idea was too make some kind of RPG however some of these ideas were deemed way too ambitious for me to complete in the time frame so I moved onto the current idea.
I decided to move forward with the 3D platformer idea, I always knew it was going to be based on Sci-Fi, but I really wanted to push that idea further than the game just being another generic sci-fi adventure. Then it hit me to make a game about a disgruntled worker from Space who is on minimum wage and is also miserable. And thus Space Caretaker For Hire! was born.
This game was a challenge to develop, I really wanted to push my skills with this project but also try to remain as realistic as possible about what I could achieve. This Postmortem dives deep into the mistakes and victories of the project and how I overcame the many different challenges that the project threw at me.
What Went Right
1. The Game Vision
The original vision of the game set out in pre-production all mostly stayed the same with the final product. Space Caretaker definitely went through changes but nothing major which changed anything from the initial pitch of the game. This is something I am really proud off because I really wanted my production time to be spent on actual production rather than having to constantly rework my vision of the game to be something that it isn't. The fact that I am able to look back at my original documentation and too see that it is still very much consistent with my final game, is something that I see as an achievement.
2. The Development Process
Generally, over the last 6 months of Production, the project's development process has been consistent and there were not any notable delays which halted the projects progress (apart from the github problem). Game mechanics were implemented at a good pace and even parts of the development cycle which I am not so confident with such as making a 3D environments stayed on track. Everything just felt pretty consistent all the way through. For the first time in a while, I didn't feel like I overworked myself to death. From the start, I always said this project was all about trying to find the right balance and I feel like for the most part, I found the balance that worked for me. All of this while still getting the results I wanted from Production. The only time this slipped was during the last month of development where I definitely worked much harder.
In regards to what skills I developed during the process, I think I played it pretty safe as not much upskilling was needed due to the work I did in Year 2, however I discovered where my weak spots are and where I need to be improving, for example, I didn't have many issues with scripting but I can get lazy with my code. Making tidier code and making better use of command patterns is something I need to do going forward. I also feel like I am in a strong position to start exploring the art side of game development, I really want to develop my skills for developing assets, painting textures and drawing.
3. Collaboration and QA
Compared to my previous projects, I definitely did more collaboration for this project and it paid off! Over the course of my third term, I worked with the Game Development Year 2 Cohort who helped with various areas of development. I had 2 students help me with 3D asset creation, which was extremely helpful and the results came out pretty good. The work ethic from them was also something that stood out to me and it was easy to communicate with them about the project.
I worked with Year 2 students on Quality Assurance too, admittedly, this collaboration had mixed results but was overall successful. I had a team of 3 for QA which over time turned into only 1 person giving constructive feedback consistently. Their feedback was valuable throughout and it did improve the overall quality of the project. I also did bulk playtesting sessions with the Year 0 students which gave me some interesting real time feedback.
Here is some of the major pieces of feedback that I got because of the QA sessions:
Floaty player controls
Gaps in the environment
General bugs and issues
Parts of the level were redesigned due to them not being fun.
I also worked with a Level Designer who assisted me in creating bitesize platforming challenges for this game. This also proved to be a successful collaboration with some solid design's being made. The only issue with this part of the collab was the lack of proper level design documentation. I also did work with an Editing and Post-Production student on Visual Effects for the game, doing this added an extra "flare" to the overall game experience with the effects really improving the game visually.
Finally, what I would call the most successful part of this whole project was my collaboration with Games & Animation Year 0 student Duck. Originally, I was going to make the 2D characters in the game myself, but my collaboration with Duck changed this after they agreed to work with me in improving the character art in the game. Not only did it save me time but Duck's work really bought my game to life, far beyond my original plans. I'm hoping to continue collaborating with Duck in the future with this game and hopefully future projects.
4. Game Visuals
I decided to go with a 2D-3D hybrid visual style for the game, with the characters being 2D and the environment being fully 3D. This was a really experiential style to go for but I feel it worked out nicely. With Duck's fantastic character designs, the characters look amazing when in the 3D environment. The environment itself turned out really well too, considering most of the landscape assets were made by myself. Once textures are applied, it really makes the world feel alive and similar to the concept art I did back in pre-production. The textures were originally drawn by myself too but I upgraded later in development to textures that I bought from the Unity asset store, this was honestly the best decision because it improved the visuals drastically.
Considering that the visuals were not my main focus, I am very happy with the results.
5. Level Design
I decided to only create 1 main level for this version of Space Caretaker For Hire. My reasoning for this was so that I could put all of my effort into one level that was really solid and fun to explore. This was ultimately the best move as designing this level took up a lot of development time. I had originally designed the level during pre-production, once I white boxed it out, I realised that 1 half of the level design didn't actually work as well as I had hoped. So, I had given myself extra time to redesign the level multiple times until I found what worked. I'm glad that I did this because the redesigned part of the level is really nicely designed now and it got positive feedback from QA. The level is now more vertical where before it was really open-ended. The new redesigned area is by far the strongest part of the level and really helps to tie everything together. I'm glad I decided to focus on Quality rather than quantity.
What Went Wrong
1. The Cost of Ambition
It's no secret at this point that I like to go the extra mile with the projects I work on, I like to push myself to see what my limits are. Space Caretaker For Hire was always a huge project from the beginning, even with the scope made smaller throughout production, the end result still ended up being a lot of gameplay. I feel I can admit that there is way too much in the end result, the one thing I keep thinking now is that sometimes Less is more. Whenever I share my game with anyone, it's always hard to show everything you can do in the level and I really wish I hadn't tried to cram literally everything into one level of what could be a full game. Most of the level feels like "busy work" rather than anything that really adds to the gameplay and experience.
If I was too do this again, I would make the experience more focused and I'd redesign parts of the level so it wasn't so open mainly the first half of the level. The mini-levels themselves would be scrapped and instead I'd work out a way to incorporate those gameplay challenges into the main level itself.
Finally, because of how much content I had, towards the end of production, it became very hard to keep track of everything that needed either bug fixes or polish. I had made life more difficult for myself and many bugs and issues are still in the game just due to a lack of time.
2. Github Issues
Throughout the whole of development, I've been using Github to backup my project and too also make it easier for me to transfer my project from Desktop to Laptop for whenever I went into Ravensbourne. For the first half of the project, I always seemed to have issues with uploading my files to Github. This was an issue that I kept ignoring and just thought it was an issue on GitHub's side.
Then around April, all of these issues came to a head and I nearly lost my whole project. This was my mid-project crisis and for a few days, it had me panicking and scrambling to get my files back. One of the worst experiences I've ever had when making a game. Fortunately I managed to get my build back with only a days worth of work lost.
After a week I then went back to Github and reuploaded the project. This time however I realised that the setup I had before was wrong and that I was uploading unnecessary files to my repository. From that point, with the correct setup, Github has worked perfectly. The lesson learnt from this is too make sure everything is setup properly with the Gitignore document before continuing with the project.
3. Character Controller
During pre-production, I decided to buy a character controller from humble bundle in the hopes that it would save me time during production. Making character controllers is a time consuming process so I wanted to save that time so I could work on other aspects of the game. This character controller for the most part works really well however as I progressed through the project, it was clear how limited I was in regards to actually extending the functionality of this controller.
The controller itself uses Unity's Rigidbody system to work but it actually overrides it by using its own systems and functionality. Meaning that I couldn't do anything with Physics with this controller. I originally had a grappling hook mechanic which worked really well but had to be cut because the controller doesn't work with Unity physics in the way that you would expect. Little issues like this led to simpler character controller than I wanted. If I had to do this process again, I would just make a custom controller, it would take more time but I could add all this extra functionality if I wanted without any issues.
4. Accessibility Features
My biggest regret with this project is that I wasn't able to get the Accessibility features implemented due to a lack of time. When I did my dissertation, I chose to do Accessibility because I wanted those features in my FMP. However, what I have discovered is that these features require more time than I thought to implement. If I had an extra month, that whole month would be needed for me to implement Accessibility and too make sure it is done right. I underestimated how long it would take to do get this done and I realistically couldn't do it.
5. Build Mechanic
The build mechanic was to be the most important part of this game and in my opinion, it didn't work out how I wanted at all. I was always back and forth on this mechanic, it went through many different iterations. The concept has always been about giving the player the option to mix and match different parts of a build but after it was done, it wasn't actually that fun to play around with. This is one major part of the game that I wish I could change and redo all over again. I also think that the UX/UI experience of this mechanic is also a reason why it didn't feel fun to use, that side of the mechanic also needs to be taken back to the drawing board and reworked.
Reflecting on Mentor Feedback
Throughout Production, I was mentored by developers from Rocksteady games. This was a really helpful experience and even though I didn't meet with my mentors as often as I would have liked, I was still able to get feedback that had a positive impact on the game and in my way of thinking. The biggest piece of feedback that I always got was about the size of my game.
This came up in every meetings and after finishing the project, I can understand what they meant. There were concerns that I was introducing too many new mechanics without focusing on what I had already implemented. My mentors were right, there was too much here and I struggled to focus on polishing everything due to the amount of features present. Another piece of feedback was about the build feature and the lack of places where you could actually build on the map.
Essentially I agree with all of their feedback and I tried my best to apply that feedback. For many of the issues they mentioned, at the time I felt I would be able to address them but I'll admit that I wasn't able too. Going forward into the future, I really need to think about the scope of different mechanics and not be afraid to admit when something isn't working whether it been functionally broken or it isn't a fun mechanic.
What I Would Do Next Time
A list of things that I would do differently if I had the chance to do it all over again:
Better build mechanics which had more variety and were used more often in the level
Redesign of Level 1
Scrap the mini-levels
Make my own custom 3D controller
Make the art style of the environment more consistent
Accessibility Features implemented
More missions
The ability to go underwater
Upgrade pathways for the player