7 years in the making?

In a way, this game has been 7 years in the making.

I always wanted to make games, but didn’t try seriously until 2018, when with some friends we founded some sort of gamedev collective to try to make something cool.

I have always been a big fan of stealth games and I played so many, but at the time I noticed that most of them focus on 3D environment, while the 2D genre is more scarce. I played a lot of great 2D games, but each time I noted a few things and details that I would do differently and over time I started to note these down.

The biggest inspiration for this game is fore sure Gunpoint and Deadbolt. Especially when it comes to the level layout. From a game design perspective it’s a very interesting layout that is not very used (if anyone knows of other games that use it, please let me know!). When I was playing both games, I realized that it would be a very good level layout for a 2D stealth game, so I started to think about it and eventually translated my thoughts and my ideas on a Game Design Document detailing how my “dream” 2D stealth game would look like.

First thing I made in 2018 was a 10 page GDD.

Then, still back in 2018, I created a first prototype for this game, very barebones but with the core gameplay loop kinda working. At the time I barely knew how to code, so it took me a good while to get it working.

It was just one level and nothing more. This level is still playable in the new version, I recreated it pixel by pixel to test if the new game felt the same as the original one.

The original prototype vs the new one

This first 2018 prototype was fun. I had fun playing it, other people had fun playing it. That motivated me to continue tinkering with it…until I stopped. Life happened. Then it happened again. Then covid happened. Then a lot more stuff. Changed cities two times, changed career, changed life. This project just became one of the many that were started and never finished.

But somehow, over all these years, the idea really stick with me. In the back of my mind it popped up from time to time. “Hey, I should finish that game prototype and publish it. It was cool!”. I think it’s rare for ideas to stick for so long, I usually have new ideas often and get super exited, only to forget about them after a few weeks. For once that an idea stuck for so much time, it must have meant something.

Fast-forward to 5 months ago, I had a rare moment of mental clarity with some free time in hands and I decided to go back at it.

Development

Initially I wanted to build on top of my old prototype, which was built in Godot 3. I thought I could save time, but actually I learned quickly that it was easier to start from scratch.

I really wanted to use Godot 4 because of this amazing little node called NavigationAgent2D. I might do a deep dive on another post on why enemy navigation was tricky to do with this level layout, but it boils down to the fact that vertical movement doesn’t really exist and every stairs is just a teleport, which can make things quite tricky and it took me a while to get right.

Problem was, migrating the project to Godot 4 was a nightmare, as most things broke. And also the original prototype was not in great shape either. I was s hobbyist developer at the time just learning to code, while now my job is software development so I learned a few things in the meantime.

Opening the Godot3 project with Godot4 was fun

This means that I rewrote the game from scratch in the last few months, trying to get the core gameplay loop as tight as I could. Free of bugs and as close as I would imagine it to a final release. Graphics will change, there would be more mechanics, more polish, a plot and worldbuilding, but I wanted the core gameplay to play and feel the same.

It took a lot of back and forth, tweaks and countless iterations to get everything right, but I’m satisfied of the current state.

Future plans

I want to make a 2D stealth game that I would enjoy to play and replay. What i did until today is a good start, but there is so much more stuff I want to experiment with.

The goal is to identify the right amount of stealth mechanics that allow to create a good challenge, interesting level design, and replayability, while trying to still keep things at a minimum. I don’t want the game to become bloated, also because realistically I wouldn’t have the time to build everything if it was too ambitious.

This game is going to be some sort game design playground in which to experiment with different mechanics to find that sweet spot. Some of these mechanics I want to test are listed on the main page of the game and I will probably make some posts about them too as there are many interesting considerations and angles on how to approach them.

For example I would like to add hostages to bee freed, a bit like Mudokons in Abe’s Abe’s Oddysee, but this would have a lot of implications in other aspects of the game.

I forgot to teach hostages how to open doors

But the core idea is to eventually have a game that can be replayed in many ways:

• a “pacifist” run where you pacify every enemy instead of killing
• the opposite, a “kill-everyone” run
• winning the game with 0 player deaths
• winning the game only killing main targets (to be added)
• etc.

This could be very fun and also very speedrunning-friendly.

In general this can be considered a somewhat hard and punishing game, like Super Meat Boy, where you need to master the core mechanics to defeat the more advanced levels. And once mechanics are mastered, it’s usually fun to use this mastery to complete the game in different ways.

Some more wild ideas that I also had, but have not seriously considered for now, is to use this experimentation about finding the right balance of 2D stealth mechanics to then build a 2d stealth roguelike/lite game. Could be a lot of fun

Another idea is about some co-op multiplayer gameplay, with asymmetric skills (for example one player can kill, the other can distract the enemies). Multiplayer stealth is very unexplored territory, which makes it exciting.

More devlogs?

Balancing such a constrained environment come with a lot of thought and considerations. For example something that looks as simple as “adding a knock-out mechanics” comes with many important questions like:

what happens if an enemy sees another knocked-out enemy?

• should it wake it up? then what happens then to both of them?
• should they panic and start to look for the player?
• If they panic, should I add a level-wide “high alert” mechanic where all enemies get alerted?
• …and if so, should I also trigger it when they see a corpse?
• …and if so, should I add a mechanic to hide the corpses?

As you can see, the rabbit hole goes deep fast. And it’s easy to find yourself in a position where by following all these thoughts, the game now requires a massive update. I would probably need to rewrite most of the enemy state machine just to tackle the first two points, let alone the rest. Might be worth it though!

I will for sure write more about this as I approach these problems, it’s very interesting to think about it.

That’s it for now, please share any feedback if you played the game!