Surviving Mid-Development Hell

It’s been a long hiatus on my developer blog as I’ve been stuck in what game developers often refer to as “Mid-Development Hell”.

It’s a phenomenon where the end is just not within sight and each day is a slog, sometimes wasted re-designing or re-coding features which was already implemented (it seemed like a good idea a few weeks ago!).

So let’s start from the beginning. My vision for the next game project would be an RPG, an open world that’s highly dynamic, and would entail a robust crafting system. The setting would be an apocalyptic & dystopian future where the few human survivors must band together to face the harsh new world.

I wanted it to be a combination of games that I really enjoyed growing up, so it would be my take on a apocalyptic Rune Factory (Harvest Moon w/ RPG combat focus) inspired game with lovely pixel art.

I’d imagined the player character building up a settlement, recruiting other survivors to join, trading with various other settlements, fighting wild mutated beasts and raiders to become a force to be reckon with. Eventually form caravan routes to trade in a big way, and leading war parties to expand territory, all the while, the small details, such as tending to crops, animal husbandry & crafting would provide some stress relief.

So it started as such, I designed the gameplay systems and mechanics, databases for items and crafting, start coding on the farm management features and proceeded to commission major pixel art.

After a few months into it, as it dawned on me the scope involved, in particular the quantity of pixel art & animations required, would result in a game that would be massively over-budget and over-time with the little resources I had available. The kind of game that I wanted to make requires a team of artists working full time, at the least, rather than a single part-time commission artist that I could barely afford.

The reality check was quick, I knew this kind of game was beyond my solo ability & resources. The scope had to change to a theme that requires less art assets thereby I could still focus on gameplay and my limited budget could afford to carry it through to completion.

I decided essentially to make a Star Nomad 2 style game (rather, no animation frames!), but down on earth, an apocalyptic sandbox, instead of spaceships, it would involve vehicle combat. It made a lot of sense to me, because I could imagine survivors and settlements in the wasteland, one that’s ruled by whoever had the biggest army of modded and weaponized vehicles. The setting takes care of the plot and dynamic world, as factions fight among each other for crucial resources. Plus, who doesn’t like Mad Max?

Such a setting allows for dynamic trade economies, roaming NPC survivors, raiders, beasts, all kinds of interesting things on the world map.

So I re-started from scratch, entirely new project codebase. I begun by working on the dynamic worldmap, the different factions, the settlements and how they interact with each other.

Because I wanted a dynamic world, little things became very important, example: How raiders spawned, not out of nowhere, but from an unhappy or resentful populace if their living conditions deteriorate forcing desperation.

People generally do not commit evil unless the situation is dire, so I designed a world that tracks each tradegoods from production to consumption. I simulated each “citizen” of this wasteland, whether their needs (food, water, meds, luxuries) are met. If their local settlement is attacked often by mutated beasts or raiders, they would grow restless and resentful of the faction that is supposed to provide their security in exchange for the goods they produce.

Raiders would grow in force over time if the militia forces cannot suppress them, as the chaos they cause, draws more desperate survivors to their side. It was a tremendous delight to watch as the simulation runs, each time different outcomes, unexpected.

Yes, a caravan hauling homebrew moonshine to the mines, if it never arrives due to beast attacks or raider (or player!) ambushes, those miners do indeed go on strike, putting a stop to the flow of crucial resources that has escalating consequences for the factions.

The worldmap as I’d imagine, and put into code works, I was very happy with this aspect of the game. So I started on the next phase, getting the car combat gameplay to work.

I had to code the car physics from scratch due to things that work in space, simply do not work on the ground. Vehicle collision needed to happen, it needs to feel visceral (screen shake!), but physic-accurate. Things like nudging a hostile car from the side rear will indeed cause it to go into an opposite spin, or all kinds of cool collisions and the weapons that enable it (steel bumpers!). It all worked, which was a surprise.

I then added weapon mounts on the cars, fixed weapons, gimbals, turrets (passengers, with RPG stats!), and got into bigger battles… it was a mess. Total chaos, with AI cars colliding everywhere, despite me coding for them to frantically steer to avoid collision (unless they are spec for ramming damage!).

Any battles with more than 4 vehicles per side degenerated into a typical bumper kart-racing at the carnivals. It just didn’t play good.

Did I mention I had already commission a lot of vehicle pixel art? Because they were done and $ spent.

But I could not continue with a real-time vehicle combat system, where cars drive around constantly, evading and attacking, not as a 2D game where the focus is on squad vs squad combat, of as many units as possible. I experimented with waves, such as battle initiates with 4 vehicles per side, and reinforcements would stream in over time.  But in the end it just lack that epic feel that an RTS in space achieves in big fleet battles.

If not real-time combat, then why not turn based? Even as a card-based tactical RPG, it could work…  especially considering one of the best games I’ve played recently, Thea: The Awakening, had a very simplistic-art card-combat system, but it manages to be tactical and engrossing.

So I experimented with a turn-base system, both as a card combat gameplay and also as a grid-based traditional turn-base RPG (Fire Emblem, Tactics Ogre, games I love).

However, there was just that voice in my head that wouldn’t go away, it kept on saying: “if it’s a turn based RPG with Mad Max style vehicles, it would just lose the biggest appeal, that is the adrenaline fuel-pumped action aspect of this genre”.

Turn base RPGs, especially with a grid, just feels better with characters rather than vehicles. Plus, as a tactical RPG, I and players most likely have a stronger attachment to their characters than a car. One that they can skill up, spec out as they wish, with a personal backstory for each character, it would also serve as an interesting point for side-quests.

To be continued…