I'm Building a Zombie RTS and the Hard Part Isn't the Zombies

Ajouté : 2026-05-27

[00:00:00]You can build something that works.

[00:00:02]Getting it to keep working is a different problem entirely.

[00:00:15]>> [music] [music] >> Last episode, I told you the objective system was coming.

[00:00:30]It's here. The orders log is displaying objectives [music] in sequence. The timer has been updated, and the architecture is in place.

[00:00:39]Progress made. But getting here cost more than I expected. [music] And what it cost taught me something about the gap between a prototype and a game.

[00:00:55]>> [music] >> The objective system is working. Open the game, and the orders log displays your first objective. Move an army unit to an urban center, complete it, and the log acknowledges it, clears it, and drops the next one.

[00:01:15]The sequence runs. The architecture is in place.

[00:01:20]The timer has been updated, too.

[00:01:22]Previously, the clock ran in hours. Now, it runs in hours and minutes. That's a small change on paper, but it does real work in practice.

[00:01:31]You can see time moving. You can feel the pressure building. 72 hours feels like a countdown, rather than a number sitting in the corner of the screen. So, the sprint delivered. The objective system exists, the timer is sharper, and the game is closer to what it needs to be.

[00:01:50]That's the good news, but here's what else the sprint delivered.

[00:02:07]>> Getting the objective system to display correctly took longer than it should have.

[00:02:12]Not because the logic was wrong, because the code underneath it wasn't built for what the game is becoming. This is a problem every developer hits and nobody talks about it enough. The code that got you to a working prototype is not the same code that gets you to a finished game. It was written fast under different assumptions solving a different problem. It got you to MVP.

[00:02:35]Now it's the thing holding you back. In my case, the objective manager wasn't communicating correctly with the display system. The architecture that worked well enough when the game was simpler started showing cracks the moment I asked it to do something more complex.

[00:02:51]Nothing was broken exactly, but nothing was clean either.

[00:02:55]And unclean code in a prototype becomes a serious drag on velocity when you're trying to build on top of it.

[00:03:03]This is the first time in this project I've felt that drag.

[00:03:07]And it won't be the last.

[00:03:19]>> [music] >> The honest version of Solo Dev isn't a straight line from idea to finished game. It's progress, then resistance, then a decision about whether to push through or step back and rebuild. This sprint, I pushed through. At some point soon, stepping back is going to be the faster option.

[00:03:54]Here's what the architecture work revealed. The objective system is in place, but the objectives themselves aren't good enough yet. The infrastructure exists. What goes inside it still needs work.

[00:04:07]A good objective in this game needs to do several things at once.

[00:04:11]It needs to teach the player something about how the systems interact.

[00:04:15]It needs to connect to the timer in a way that creates genuine pressure.

[00:04:20]>> [music] >> And it needs to drive the game forward rather than just filling time.

[00:04:25]Right now, the objectives are functional placeholders.

[00:04:30]They work as a demonstration that the system runs. They don't yet work as a game.

[00:04:36]The gap between a working system and a system worth playing is a game design problem, not a coding problem. And that's where the next sprint is going.

[00:04:45]Not more architecture, not more features.

[00:04:48]A focused effort on what the objectives actually ask the player to do, why it matters, and what happens when they get it wrong.

[00:05:02]The spiral mechanic I talked about in episode one is the answer to the last question.

[00:05:07]More infections breed more infections.

[00:05:10]More refugees breed unrest. Army presence stabilizes both.

[00:05:15]The objectives need to connect to that spiral in a way that makes the player feel the consequences of every decision.

[00:05:23]That's the work ahead.

[00:05:25]The architecture can support it.

[00:05:27]Now it needs something worth supporting.

[00:05:31]One question before you go, have you ever hit that wall in your own projects?

[00:05:37]Where the code that got you started becomes the thing holding you back.

[00:05:41]How did you handle it? Did you push through or step back and rebuild?

[00:05:46]I genuinely like to know.

[00:05:48]So, leave it in the comments.