Game Devlog: Eight Weeks of Progress
It’s now been eight weeks since I decided to work on my own projects full-time.
The plan was to spend around two months on a small and simple project to get experience with taking a game from start (conception) to finish (selling on Steam). For reasons that will be explained later, that has turned out to not be feasible (as probably anyone who’s ever shipped a game could have predicted).
A month ago I posted a video of my progress to Reddit and was surprised to see that it got a lot of positive feedback. I expected maybe 30 upvotes and a comment or two. Instead it received 500+ upvotes and 50+ comments.
Many people complimented the visuals which I didn’t expect because I’m not an artist. Some also complimented the sound design which, again, was unexpected.
A few comments offered some constructive criticism which I took seriously, the main one being that the horror feel was well done but the powerup mode didn’t fit the mood or aesthetic. I agree.
I set out to create a game that took the overall premise of Pac-Man (wandering around a maze, collecting some items, avoiding four enemies, getting a powerup that let you turn the tables on the enemies) but from a first-person perspective. Somewhere along the way the game became darker and moodier and more horror-esque.
The fact that I accidentally created a horror game probably reveals something about my personality, but regardless of intent, I seem to have stumbled upon an aesthetic that some people find compelling.
So I decided to lean more into the horror aspect by maintaining the oppressive mood.
I realized I needed to move away from the Pac-Man gameplay element of a powerup that made the enemies run from you, so that meant removing the ability to attack the skulls and the accompanying upbeat music.
To accomplish that, I changed the powerup mode to instead make the walls transparent for a brief time so that you could see what coins you were missing. You’re unable to see the skulls through the walls, and they no longer fear you, so it ensures they remain threatening throughout.
I like diegetic user interfaces where the UI information is placed in the game rather than pasted on top. My favorite example is Dead Space, where your health is displayed on the back of your suit. I think Dead Space demonstrates that, especially for horror games, a lack of a traditional UI can help maintain a game’s atmosphere.
My solution was to create a central safe zone in the middle of the maze that holds the powerups (so that you would be able to grab one and have a 360 view of the entire maze) as well as a coin indicator that displays how many coins are left to collect.
I’m still not satisfied with this solution though because while the score is technically “in the world”, there’s no logical reason why there would be a floating coin with a number hovering over the top. Nonetheless, it works for now at least.
To further complement the mood I added:
- Footstep sound effects
- An idle animation
- Head bob while moving (configurable for those that hate it)
- Lowered the walls and surrounded the maze in a swirling tempest of doom
I modified the audio system to be able to have attenuated positional sound sources to enhance the soundscape. Right now it’s just the Left and Right channel, but maybe in the future I’ll add support for rear as well.
To make the skulls even more frightening, I made it so that their mouths open as they get closer to you, as if they were going to swallow you.
A lot of unseen work went into engine features that facilitated features or simply made my life easier.
I try to write the simplest code first that does whatever I need in the shortest amount of time. Then, if the need arises later, I rewrite. I think that’s preferable to spending time writing code to be “future-proof” that ends up never being needed.
Some engine systems I worked on were: positional audio, object culling, better asset management, shader reloading, a UI system, and scene hierarchies.
Here’s a short video of everything so far:
I wish I had started on the business things two months ago because I didn’t realize how long and involved it would all be.
To create a store page on Steam you need to register with Steamworks. The first thing they ask you is what your legal business name attached to your banking documents are. Many people recommend that you form an LLC rather than sell as a Sole Proprietership.
Creating an LLC is easy enough (although expensive in California), but then your name and address are in the public record so anyone can look them up. California doesn’t allow you to use a PO Box so you have to use a Registered Agent service and then their information goes in the public record. The one I chose is $125/yr. The LLC formation process takes a few weeks to process.
Steam also wants your bank account information. If you’re using an LLC, then your LLC’s name needs to be on the bank account. That means you need to get a business checking account, but you can’t do that until the LLC is fully formed because they require a copy of the Articles of Organization.
I still haven’t signed up for Steamworks because I’m still waiting for the LLC process to finalize.
Long story short, the process looks like this: Registered Agent (instant) -> LLC (2-3 weeks) -> Bank Account (3-5 days) -> Steamworks (1-5 days)
And it takes a few weeks or more and so it should be started as early as possible.
I’ve noticed that my highest highs are on Fridays and my lowest lows are on Mondays. I think this is because on Friday I can look back on the week and see all that I accomplished, but on Monday I had the weekend to overthink everything and begin to doubt myself.
Unfortunately this means that I’m grumpy on Mondays and am half as productive because I think everything is shit. By Friday I’m feeling good but then the weekend hits and the cycle repeats itself.
Maintaining motivation is difficult when you have your nose in a thing every day and can only ever see the flaws. I have immense respect for anyone who has spent years on the same project and come out the other side.
The Game Isn’t Fun
The game in its current form isn’t very fun.
Walking around a maze collecting coins is cool at first but gets boring quickly. I could add a timer so that the goal is to collect all of the coins as quick as possible. Then successive runs would be attempts at finding the best path to collect everything in the shortest amount of time.
But the enemies aren’t dangerous and essentially add nothing to the gameplay. They’re creepy, sure, but dodging them is easy because if I make them faster and smarter then the game becomes frustrating.
The original Pac-Man worked because you had all of the information at a glance: Pac-Man’s location, the pellet locations, and the ghost locations. Good players learned the enemy behavior and were able to predict their moves and avoid them.
That doesn’t work in first-person because you never know if there’s a skull right around the corner. That’s what makes it tense and creepy. But if I make the skulls faster and smarter then the player is constantly being cornered and caught which doesn’t feel fair because you had no defense against it.
Finding the Fun
I need to take some time and figure out how I can make the game more fun without throwing everything away.
Maybe that means adding additional hazards, items, enemies, and mechanics.
Or maybe it means moving away from the Pac-Man inspiration entirely and finding something else that suits the atmosphere.
Last Edited: Dec 20, 2022