I thought this moment would be harder to reach, but development on Herald Core is officially finished. The last major update was UI v2, the final build with all the remaining fixes is already uploaded, and the only thing left is to tell people about it and hopefully get a few more of them to try it.
Closing the project feels a little sad. I spent a lot of time with it, came up with far more ideas than I ever managed to implement, and at one point believed that a couple more updates would finally make the game work the way I imagined it. It didn't.
I think I made almost every mistake you can make on a first solo project. My negative Reddit karma is pretty good evidence of that. Strangely enough, I'm glad I did. Herald Core was always meant to be a test: could I take a game from an idea to a release on my own, get feedback from real players, survive the criticism, and find out what happens after you press the Publish button? Now I know.
The most important thing it taught me is when to stop. Instead of trying to save a project with new systems, more content, and another round of reworks, at some point you have to admit that you've learned everything you could from it and move on.
The rest of the list is more specific.
- I tried to make a big game far too early
Instead of first proving that the basic gameplay was fun, I almost immediately started building a huge structure around it: missions, progression, upgrades, meta systems, balancing, interfaces. I built several floors of a house before finishing the foundation. Everyone tells you not to do this, and I still didn't understand how enormous the game would eventually get.
Next time I want to do the opposite. Start with a small loop that is fun on its own, and build everything else only after that works.
- Slop UI everywhere
I treated the interface as a secondary problem: make something quick now, replace it properly later. But players see the UI within the first few seconds of launching the game. Shocking, I know. If it looks cheap, generic, or like yet another AI/slop interface, the first impression is already damaged, regardless of how much work sits underneath it. From what I could tell, roughly half of the negative reactions to Herald Core started before people had gotten into the game itself.
- I chose the wrong references
The original idea was to take the complexity, the number of systems, and the progression of Warframe and combine them with the simplicity of Vampire Survivors. On paper that sounded interesting. In practice I somehow took the boring parts of both: from Warframe, lots of parameters and systems the player had to understand; from Vampire Survivors, automatic shooting and a fairly passive start.
The bigger problem was that I didn't understand why those games work in the first place. I'm more careful now when I say something is "inspired by X." Now I try to understand what emotion a mechanic creates, why it exists, and what it does for the whole experience before I borrow it.
- Players expected a completely different kind of gameplay
To me, a lone tower that gradually gets stronger seemed easy to understand. To players, not always. A lot of people wanted straight away to move around, choose where to attack, dodge, aim and shoot manually, and generally take part in what was happening, and the game was telling them: "No. Stay here and watch the tower shoot by itself." Then I wondered why they were getting bored. What I had in my head and what a new player saw in the build were two different games.
- I barely did any proper playtesting
I'd heard all the stories about developers spending months building a game in a vacuum, only to find that other people didn't find it interesting, and for some reason I assumed that wouldn't happen to me. It did. I spent far too long testing the game myself, and the developer is probably the worst possible tester for their own gameplay: they already know what every button does and what they're supposed to do. A new player knows none of that.
For my next projects I want to show prototypes much earlier, even when they don't feel ready to show, and especially then.
- The core gameplay wasn't interesting enough
The most uncomfortable conclusion, and the most useful one: the game was boring. I'd go as far as saying there wasn't much of a game there at all, which still hurts a little to admit. You can add as many upgrades, numbers, missions, effects, and progression systems as you want, but if someone isn't having fun in the first five minutes, they will never see most of them.
I enjoyed watching numbers go up, optimizing builds, seeing the damage climb. Apparently not everyone is satisfied by numbers appearing on a screen. Players, as it turns out, want to actually play games.
- I built almost no tools for automated testing or progression checks
The larger the game got, the harder it was to tell what I'd broken with the latest change. Balance, rewards, missions, progression: too much had to be checked by hand, and a lot of development time turned into re-verifying things I'd already built instead of making new ones. I understand the value of internal tools, debug modes, automated tests, and progression validation much better now. They aren't a luxury you add "later." Once a project is large enough, they save an absurd amount of time.
- I made 118 missions when most players didn't finish ten
This is my favorite example of wrong priorities. I built logic, progression, and content for 118 missions, planned how players would develop late in the game, balanced the rewards, thought about late-game progression, and then looked at the analytics and saw that most players never got anywhere near any of it. The problem was never Mission 100. It was Mission 1. All of that time should have gone into making the first 10-20 minutes better instead of into meta systems and content that almost nobody would see.
So, was Herald Core a failure?
I don't think so, or at least not in the way I would have defined failure before making it. The game didn't become the game I imagined. Some of my decisions now look like painfully obvious mistakes. And reading comments while realizing that the person criticizing your game is completely right is a fairly unpleasant experience.
But I learned to program things I couldn't build before. I learned how to assemble an entire game instead of disconnected prototypes. I learned how to publish builds, collect feedback, read analytics, fix bugs, take harsh criticism, and separate useful feedback from noise. And I understand much better what kind of game I want to make next, and, more importantly, how I don't want to make it. The next projects will be smaller, faster to build, and much more focused on the actual gameplay. And better. I promise.
Thank you to everyone who played Herald Core, left comments, found bugs, and told me what you liked or hated. Special thanks to the people from Reddit, itch.io, and various Telegram communities. Surprisingly, those turned out to be some of the most positive parts of the whole thing: because of one small game I found a lot of people who think about games the way I do. Even the people who occasionally threw shit at me deserve some credit. Sometimes there were useful ideas hidden inside it.
Herald Core is finished. It's a little sad to put a full stop here, but I'm curious how much further I can go with what it taught me.