Core Engine Vertical Slice & Lessons from End-to-End Testing

10/2/2026

There isn’t much I can share this time. I’ve been quite busy, so I only made limited progress on this game dev log. Still, regardless of how big or small the progress is, I will continue to document it on this blog.

This time, I focused on creating a vertical slice for the core engine. Previously, during development, I mainly relied on unit testing. However, for this stage, I conducted end-to-end testing instead. Since I haven’t established proper testing standards yet, I often found myself confused when the test results didn’t match my expectations.

Because of that, I decided to write down some key points so that future vertical slices won’t become as confusing:

• For an event-driven engine, testing should be designed to simulate real player behavior as closely as possible.
• Be careful when executing code synchronously, because in real gameplay, player input naturally involves delays. Therefore, it’s important to introduce artificial delays or input queues to better mimic actual conditions.
• Running a full vertical slice (for example, simulating 100 in-game days consecutively) with limited testing tools may not produce optimal results. Constantly tweaking the code or even modifying the core engine can easily become time-consuming or lead to over-engineering. Moreover, some of this code may be discarded later when the project moves to the next stage.
• Testing should follow the principle of keeping things simple while still covering all core engine components and their interactions.

That’s all for now. Next, I plan to start working on a simple UI for the game.