Overlooker is a 2D top down gameboy color style horror game. My main goal in making it was to learn about the general process of game development, and if I had a creative goal it was that it might hopefully manage to spook somebody. I am a huge fan of old Resident Evil and Silent Hill style survival horror games, and I figured that if I could creep someone out with a small gameboy color game I might have some potential as a survival horror game developer. I expected nobody to view the game, play the game, or to receive any donations from it. I developed the game in 22 days (although I was sick for two of them), and had never used GameMaker Studio or made art assets prior to making the game. Overlooker was released for free (Pay By Donation) on January 2nd, 2015. You can download it from itch.io!
What went right:
1 – Success!
I managed to be successful on every level.
To date Overlooker has had roughly 8,000 total views, roughly 1,500 downloads, and made $40 through donations (the thought that multiple people actually went out of their way to donate and support my work is very heartwarming). While these numbers may not seem like much, I consider them success since I expected nothing. I also did not promote the game at all outside of posting on my social media pages, and sending messages to friends through Facebook. I knew the game was not good enough to send to news and review sites, so I chose not to do so. That makes me very happy about the amount of downloads (I don’t have 1,500 friends!)
There are also about 30 Let’s Plays of the game on Youtube, which has been very helpful in learning what went right and what went wrong. Most importantly, some people (either through e-mails or from watching Let’s Plays) actually seem to have gotten creeped out at some point playing the game. This is what I consider the major success.
2 – Narrowing down the scope
My very first music demo 10 years ago was a three track black metal album. The riffs were simple, the beats were simple, and it was short. But the amount I learned from going from start to finish was massive, and I think you can only learn some of those things from the experience itself. Each album you make, you learn something new that you end up applying to your next album to hopefully make it better than the last. There’s so much knowledge in every field, that there’s always room to learn and improve.
I made the decision to limit Overlooker’s development to 22 days for two reasons; one, because I made it during winter break between semesters and wanted to finish it before returning to University, and two, because I wanted to actually finish and release something. Since my goal was to learn, I didn’t worry about how bad certain aspects of the game might be.
I narrowed down the scope of the game considerably. I originally wanted to have a Resident Evil style inventory, a save system, a complex memo/document system, but I cut all of it. The game is as simple as it could possibly be, and being able to cut down things that didn’t necessarily water down the point of the game itself was important. It helped me focus on learning the overall process of completing and releasing a game in 20 days rather then fiddling with a chunk of inventory code that won’t work for too long.
3 – Music
The soundtrack has 12 total tracks. 4 of them were taken from previous solo albums of mine, while the remaining 8 were written and recorded in less than a day. Considering the large amount of positive feedback I got about how creepy and dark the soundtrack is, I consider the music something that went right! (Listen to the soundtrack here)
If I added up the amount of time spent working on art assets, it would equal about half a days work.
4 – Not making my own game engine
For some reason, many beginner game development tutorials recommend people learn C++/OpenGL and make their own game engine entirely from scratch. There are many people with the belief that if you don’t make your own game engine, you didn’t really make your own game. I find this bizarre, as I would never say someone isn’t really a guitarist if they didn’t build their own guitar, or that someone didn’t really record their own album if they didn’t build their own version of FL Studio or Protools from scratch.
I think that most people interested in game development should use GameMaker Studio, Construct 2 or Unity to build a basic game and learn about the process. Then, you can take what you learned and put it towards eventually making your dream game. While I have previous experience with C++/SFML and Lua/Corona, I chose to spend my time working on the game itself instead of the engine. I was able to learn much more about how to make a game overall as opposed to lower level things like how to draw triangles to a screen. Don’t get me wrong, I have nothing against developing your own engine; I just think that beginners learn a lot more about the overall field of game development by using an engine instead of making one. I actually think that all people involved in game development would benefit from learning to program and understanding how the technology works, not just the programmers themselves.
5 – Controls
I didn’t have anyone test the game for me. As a result, the controls are pretty terrible. I made up the controls on the fly and since I got used to them, I considered them acceptable. The majority of the complaints about the game are about the controls. I would have had people test the game, but I ran out of time.
6 – Poor combat feedback
When using a weapon, all you hear is a sound effect. There is no animation (or light from a gunshot) to indicate that you just attacked an enemy. Many people found this frustrating and they were unsure if the weapons were working properly. There is also a glitch where sometimes the character won’t go into the attack stance even though they are, and that causes even greater confusion.
7 – The game is both way too easy and way too hard
Some people found the game painfully difficult and frustrating (and the lack of a save system didn’t help) while others breezed through the game without even having to use health items. This relates to the previous two problems; I should have had people test it if I had time.
8 – No save system
I included save rooms in the game (complete with a relaxing Resident Evil style save theme) but I didn’t have time to implement a save system. Multiple people felt frustrated if they had to restart the game after dying, and some felt like they were missing the save system itself even though they had searched throughout the save rooms. I should have made it clear somehow that there was no save system.
I spent roughly 20 of the days working purely on game content. Placing enemies, placing items, placing progression events (putting a book on a bookshelf causes a new area to be accessible, etc). Based on the average length of Let’s Plays, most people get through the game in roughly 20 minutes.
This means that for every day’s worth of work on the game, the result is 1 minute of gameplay.
Keep in mind that before I started working on gameplay content, the following were already completed: Art assets, Music assets, Game Engine (used GameMaker: Studio), and Design Document. Also keep in mind that I did not spend any time doing QA/testing (and spent little time testing it myself). Overlooker is an extremely simple game and I made it entirely by myself (so there was no time spent communicating with team members). Given all these factors, I imagine that this 1:1 ratio is very, very optimistic for larger projects.
However, this does put things into perspective. If you want to make an 8 hour game, that’s 480 days worth of work, based on the ratio. This made it much easier for me to decide what my second game would entail; I kept cutting out ideas that were too large since I knew I probably wouldn’t end up finishing them anytime soon. I have been able to decide on my next game and I believe I’ve been able to make the scope of it much more realistic based on this ratio. I wouldn’t mind spending years in the future making my dream fixed camera tank control survival horror game, but I need a lot more experience before I attempt that.
There are obviously an immense amount of variables that would effect this ratio, and many more that I didn’t deal with that would effect it even more (multiple platform releases, QA, dealing with much more complicated systems (AI, Inventory, Player and Enemy stats, etc)). I do find the ratio very helpful though.
That’s the end of my postmortem for Overlooker! Although there are tons of problems with the game, I am very satisfied with my first attempt at trying to make a game. I also think I managed to do a decent job considering I only had 22 days, and had never tried GameMaker Studio or tried making art assets before.
I hope some of this information may be useful to beginning developers!