With Hex Sweeper and PL4X behind me I finally had my head around the Phaser fundamentals, and it was time to build something more ambitious. Something original. Since the long-term goal has always been turn-based strategy games, the idea was to take a familiar genre and see if it would work in a turn-based format. I kicked around a few concepts before settling on desktop tower defense. The idea was simple, waves come in turns, you get a hero with a move and an attack, and you place and control towers to handle incoming enemies.
Like most projects, the basics came together quite quickly. I had the core engine and views working within a few weeks. But it's all the little details outside of the core coding that started to drag on. Game balancing itself took an unexpected few days alone to tweak and test play. How long should a run last? How fast should the difficulty scale? I got things to a decent spot but there's still room to tune it further.
Graphics were a whole other challenge. I'm not an artist, and doing even simple visuals took far more time than I had planned for. I cobbled together a mix of AI generated images and tweaked vector icons just to get something functional on screen. I was happy anything was moving and functioning at all but it was definitely looking amateurish and half baked.
But, with something functional, I decided to release the game early, thinking I'd get some useful feedback. I posted to a handful of Discord servers and subreddits, but didn't get as much feedback as I was hoping for. Plenty of other channels to try, but I stepped back and realized the game probably went out a bit too early. It needed a graphical polish and better mobile support before any real promotional or feedback push would be worth it.
The first thing I noticed looking at the game with fresh eyes was that it was just ugly. And considering how much time I'd already poured into graphics myself, it was clear I needed to bring in help. I went with Fiverr, and it turned out to be a great call. Cost effective, fast turnaround, and the results were so far ahead of what I was producing on my own. It freed me up to spend my time where I actually add value, which is the code.
I did take on the tile graphics myself since they're simple geometric shapes, more about color and style than illustration. I added a bunch of new tile types to break up the visual monotony. Instead of a single grass tile repeating everywhere, the map now has grass, trees, plains, mountains, and water. The mountains and water also serve as randomly placed impassable tiles, which adds variety to each run and forces you to think more carefully about tower placement.
I also went back and refactored some parts of the code that I originally planned to skip until the next game. That tied into the bigger lesson I took away from this whole process. I had been trying to move fast with the strategy of getting multiple games out quickly, but seeing what other hobby devs are producing on the game dev subreddits put things in perspective. The bar is genuinely high. Shipping an old school DTD with rough graphics just isn't going to cut it. It's better to take my time, polish this one game until it's solid, get the graphics right, ship mobile support, maybe prep a Steam version, and have something I'm actually proud of before moving on.
The initial release is out and playable, and the graphics overhaul is nearly finished. Even just the new visuals alone have completely changed how the game feels. Still on the list is ripping out the old frame and going full screen with a floating menu, plus adding proper mobile support with pan and zoom. Desktop and tablet will still play with a full zoomed-out overview, but having mobile as an option opens things up. Plus, everything I build for mobile here carries over into PL4X, which I was planning to add it to anyway. Better to work it out in a live game where it gets real use than to bolt it on later.
I'm really enjoying the whole game dev process, from the coding to the art direction to figuring out promotion, and I'm looking forward to getting Button Breach polished up for a more proper release in the next few months.