Alright, so I figured I’d share a bit about this “lethal race game” project I’ve been tinkering with. It all started, as these things often do, with a thought: “Wouldn’t it be cool to have a racing game where you can just blow stuff up?” Seemed like a solid gold idea at the time. How hard could it be, right?
Getting Started – The Usual Optimism
First things first, I needed an engine. I went with Unity. Seen a ton of folks using it, and the Asset Store looked like a treasure trove. Well, it is, but it’s also a rabbit hole. I probably spent the first week just downloading “essential” assets, most of which I never even ended up using. Classic.
Then, getting a basic car to just move around. You’d think that’d be straightforward. Nope. I wrestled with wheel colliders for what felt like an eternity. Getting them to feel even remotely like a car and not a hovercraft with a drinking problem was a battle. Honestly, sometimes I think those things are just there to test your patience. I tweaked friction settings, suspension, center of mass – you name it, I poked it.
Making Things Go Boom (Intentionally)
Once the cars were somewhat drivable, it was time for the “lethal” bit. This was supposed to be the fun part. And it was, mixed with a healthy dose of “why isn’t this working?!”. I started with some basic projectile stuff. My first iteration had missiles that were more likely to hit a random seagull than another car. And then the explosions. Oh boy. Finding a balance between “cool looking boom” and “my computer is now on fire” was tricky. Particle systems are powerful, but man, they can bog things down if you’re not careful.
- Weapon aiming? A whole saga. Auto-aim, no-aim, leading targets… each came with its own set of headaches.
- Collision detection was another beast. Is it a trigger? Is it a collider? Did it hit the shield or the main body? So many edge cases.
Tracks and AI – The Real Test of Sanity
Then I moved on to track design. I thought, “I’ll just sculpt some cool terrain, lay down a path.” Easier said than done. Making a track that’s actually fun to drive on, with good flow and challenging corners, is an art form. My first few attempts looked like spaghetti thrown at a wall. And then getting the AI to navigate these monstrosities? Unity’s NavMesh system and I had some serious disagreements. It’d bake a path that looked perfectly fine, then the AI cars would just decide to take a scenic route through a mountain.
The AI racers themselves were a whole other level of “fun”. I cobbled together some state machines for them – chase, evade, use weapon. Sounds simple. But getting them to behave intelligently, to not just crash into each other constantly, or to use their weapons without wiping out the entire field (including themselves) in the first ten seconds… let’s just say there was a lot of watching them do incredibly dumb things. Lots of tweaking variables that seemed to do nothing, then suddenly do too much.
The “Polish” Phase (aka When You’re Tired of Fixing Bugs)
Eventually, it started to vaguely resemble a game. I threw in some sound effects – mostly free ones I found scattered around the internet. Some engine roars, some explosions, the satisfying thud of a missile hitting its mark. The UI was pretty basic, just enough to show your speed, lap, and maybe what weapon you had selected. By this point, my ambition for a super-slick interface had dwindled to “does it work without crashing?”
Testing was… repetitive. Driving the same track over and over, trying to trigger that one bug that only happens when you’re in third place on a Tuesday. You know how it is.
So, What’s the Takeaway?
Looking back, it was a proper slog. There were definitely moments I considered just formatting the drive and taking up birdwatching. But, seeing those little cars zoom around, blowing each other up, there’s something to that. It’s a janky, unbalanced mess in some places, but it’s my janky, unbalanced mess. I learned a heck of a lot, mostly about how much I still don’t know. And I confirmed that game development is about 10% cool ideas and 90% figuring out why that one line of code is making everything explode. Or not explode, when it really, really should. Would I do it again? Probably. Ask me after a few more cups of coffee.
