I never realized it before, but using UMG feels like handling a ticking time bomb. I don’t know about now, but back in the day, UI work was often assigned to junior developers. If you’re using UMG, that’s definitely a bad idea. UMG is great for quickly and easily getting results, but if you don’t manage objects properly, it can end up being the thing that crashes your game. I vaguely remember reading something in Unreal Engine’s profiler documentation about UI being a performance issue, though I’m not sure if that was directly related. Either way, after dealing with it firsthand, I’ve come to respect UMG’s dangers once again. If you don’t know what you’re doing, like I didn’t, it can be a total disaster. Fixing this mess has been exhausting. I’d love to rebuild the UI from scratch, but there’s already too much in place.
While working on controller support, I also wrapped up additional camera auto-adjustments and other necessary features specifically for gamepad controls. I made some adjustments to the save system as well, ensuring it can store progress across multiple persistent levels for the upcoming stages. I’ve been thinking about how to push the cassette futurism aesthetic further, but I’m not quite sure what to focus on. I even added a screen hood to the kiosks on purpose, but it barely makes a difference. There are still a lot of areas in the game that feel lacking, but it’s hard to pinpoint exactly where. That’s why I’m planning to start a closed user test in March to gather feedback. In hindsight, maybe I should’ve done user testing before pitching.