Dev Log 3: Year Two

This is the third development log on my project, a top-down twin-stick shooter mecha action game with linear level progression. 

On March 1st, 2024 I started development on Mechdivers, a solo project that I would aim to spend roughly a year to 18 months of development on and would serve as my ultimate portfolio piece as well as my first commercial Steam release. I planned to do everything entirely myself, from the design to the art, all programming and of course, the sound design and music. Today as of writing, it has been 393 days since I began working on it, and I have to say I’m incredibly proud of the work I’ve done. I’m as always very excited to share what I’ve been working on, but first let’s do a quick retrospective of how far the project has come. 

A Look Back at Year One

Around mid-2023, I played Resident Evil 2’s remake for the first time and with it being one of the first survival horror games I properly got into, I began development on my own very small scoped survival horror game. The rough idea was to take after RE2 and Dead Space, with a third-person shooter set in a sci-fi nuclear bunker, and I built quite a bit actually for the basic gameplay. I set up a third-person character controller, a modular weapon system, and a surprisingly decent stealth system. A few months in, I decided that even with the limited scope, I’d be better off with a much, much smaller scoped project that could still showcase what I could do. I initially took inspiration from Armored Core’s aesthetic and the mission design of Helldivers and got started, settling on a top-down perspective and opting for a linear game structure due to my love of linear campaign modes, finding that it helps with replayability in comparison to open worlds or hub worlds. I focused hard on a narrow scope, only a handful of missions, and a highly replayable design, seeing as that could vastly increase the value without blowing up the scope to dozens and dozens of levels. After a few weeks I had something playable, and after a few months I had an experience that could go from a menu through a short level, and back to the menu. 

Level design and art direction were things that I knew what I wanted, but couldn’t exactly execute how I envisioned. The 3D art I made was pretty rudimentary at first, mostly just blocks and cylinders to start, but modelling and texturing has unfortunately never been my strong suit. Thankfully, level design clicked pretty well, so despite only having one full level as of now, the levels I have currently planned work from the many things I’ve learned through making this first one fun and more than a minute long. I took the rough approach from the weapons system I had made previously and expanded on it, made basic enemy AI, improved on the character controller, improved models, and added Steam features, which was a big step for me. 

As of the last dev log at the end of last year, the vertical slice was wrapping up, and I was going into 2025 with high hopes for what was to come. The game was looking better and better every day, sound was finally starting to get added, and getting the game ready for marketing and eventually Steam Next Fest. 2025 would be the year that I release my first commercial game. Without spoiling too much, I’m happy to say that it might need to be extended to early 2026, but for great reasons. Now, let’s get into what I’ve done the last three months.

Year Two

The biggest and most noticeable update to the game I’ve made has been doing an art pass of the level, with this gif really showcasing just how much of an upgrade it’s been to add models. While it’s definitely not final, shippable art, it’s a massive improvement to the feel of the game. Without getting too into it for the sake of brevity, I made a lot of floor pieces, multiple wall types, three rocks to use and reuse for the terrain details, and I swapped out the ProBuilder objects of the two halves of the terrain for proper Unity terrain with textures for rock, dirt, and grass.

What I’d argue are the second biggest updates to the game are the menus and full implementation of the settings menu, something I wanted to prioritize to add to the polish early on. All of the menus got fully reworked on the scripting side, being significantly cleaner than before, but also being a bit more smooth to navigate, unless I’m imagining it. Within the settings menu, another big addition was French localization. I decided fairly early that voice-over probably wouldn’t be the best idea, being a quick way to balloon the scope, so by extension, keeping everything as text makes it a lot easier to then add support for different languages. This is going to be an ongoing process, but having the descriptions for the options both in English and French is a great step forward, and while maybe not expanding the audience a ton, it's a great experience as I’m still learning the language.

Speaking of ongoing, playtesting has been happening frequently, not only to test the mechanics and flow of the level, but to also check the gameplay balance. The ammo economy I had set up previously was just enough to get through the level; however, it was very tight, and relied on the player finding the secrets to replenish armour and ammo. As a solution, I added many more pickups throughout the level for the player, and the more fun solution, I added a “melee” attack that damages enemies and destructible objects when boosting into them. It uses three sphere colliders in the direction that the player is moving, smoothly changing position, thus the range of the melee attack, depending on the velocity of the player. This was a design choice I wanted to add from the beginning due to limited ammo leading to soft-lock potential, and I might make it so that melee kills increase the chances of item drops from enemies, but we’ll see when the item drop refactor happens.

Now for sound: it’s starting. Nearly all of the UI interactions have associated events, with only the play game, weapon selection, and secret found cues left. The big green sphere collider in the above screenshot is the new bullet flyby handler, which checks if a bullet is passing close to the player, and if it is, it plays one of 32 sound effects of a bullet whizzing by the player’s mech. With the addition of the gunshot sounds and the impacts in the future, all of that will really make the fights feel more visceral.

The ambience is still in progress, but it required a few scripts for the river in particular. To get the sound to follow the player while still being localized both on the bridge and on the cliff beside it, I needed two separate solutions to execute it properly. The first, for the bridge, would be to have a trigger volume, playing the sound, which then uses the “SoundFollowPlayer” script I created, which moves it to the closest point to the player within a bounding box, so it spatializes correctly with the indoor areas on either side of the bridge. While this could work for the cliffside, it’s a fairly irregular shape so it wouldn’t have been very clean, so I opted for using a spline. This works effectively the same way, finding the nearest point to the player along the spline, and moving the sound source there. Both of these are great and work excellently, so these are absolutely being used in the future for other levels.

Some relatively small updates are that I fixed several bugs and issues, optimizing dozens of scripts, and ultimately made things easier to read.

Production Updates

Now if you were particularly eagle-eyed, you’ll have noticed that I very carefully avoided the term “solo project” except for referring to it in the past-tense, and that was because I decided that for the good of the game I’d get an artist to make the game more visually appealing than I ever could on my own. A good friend of mine was just wrapping up on his studio’s project and is nearly done teaching for the current school year, so I asked if he’d be interested and thankfully he was enthusiastic. He’s only just started and everything already looks thousands of times better than I had it before, so I’m sure the next dev log will be mostly screenshots showing off his work; stay tuned for that in a few months. With him managing the art, that leaves a lot of time open for me to make stronger and potentially more levels, with the project possibly growing to a much bigger scope now that the visuals are much more indicative of a promising game. 

Alongside the doubling of the size of the team, I made the decision earlier this month to return to school for a postgraduate sound design and production program. It’s still ultimately my goal to release this fall, but with the year of schooling starting in September, I think it’s much more reasonable to push it back and utilize what I learn to make the sound design in particular even better as we get the game ready to ship.

Signing Off

With the first year of development done, I’m thrilled about the amount that I’ve accomplished in my spare time here. Not needing to worry about the art is such a massive weight off my shoulders for the project, and means that not only can I focus more on making the game as good as it can be in other areas, but that it will be much more marketable with more than the middling sound designer 3D art that I had until now. I’m very excited to see what the next dev log will look like, and will very likely have enough to have set up the Steam page to begin marketing. Now I need to figure out a name for it. See you in a few months!