Tenlach Arcade
In DevelopmentMarch 1, 2025
Tenlach without multiplayer — strictly level-based, high-score-driven, how far can you get through the gauntlet. The version that should have been built first every single time. A discussion of why starting single-player and adding multiplayer later is almost always the right sequencing.
Purpose
Strip Tenlach down to its core: grid combat, Commandments, worm enemies, escalating difficulty — but no networking, no matchmaking, no lobbies, no rollback netcode. Just you vs. the gauntlet. How many levels can you survive? What is your high score? This is the version that proves the game is fun before investing in multiplayer infrastructure.
Stack
What I Learned
- Every multiplayer game should begin as a single-player game. The reason is diagnostic: if the core game loop is not fun when you play alone against AI, adding another human will not fix it. Multiplayer amplifies fun — it does not create it. If Tenlach's grid combat, Commandment system, and enemy variety are compelling solo, they will be electric multiplayer. If they are boring solo, multiplayer just adds latency to boredom.
- Starting single-player eliminates the hardest technical problems from the initial build: no networking, no state synchronization, no rollback netcode, no matchmaking, no lobby management, no disconnection handling, no anti-cheat. Each of these is a project unto itself. By removing them all, you can focus entirely on the game design, the feel, the balance, and the fun. Those are the problems worth solving first.
- A gauntlet/arcade mode is the simplest progression system: level 1 has weak worms, level 2 has stronger worms, level 10 introduces a new enemy type, level 20 introduces a mini-boss. The player's score is how far they get. Leaderboards (even local, on-device) create replayability through competition with yourself. No narrative needed, no save system needed, no world state needed. Just: fight, survive, go further.
- The high-score model provides the clearest feedback on game balance. If every tester reaches level 50 easily, the difficulty curve is too flat. If nobody gets past level 5, it is too steep. If everyone plateaus at level 12, that is where the difficulty spike lives. High scores are not just player motivation — they are balance telemetry.
- Building arcade mode first also produces a playable game faster. Tenlach Arcade could ship to the App Store as a standalone product while multiplayer Tenlach continues development. The arcade version generates feedback (is the combat fun? are the Commandments balanced? do the worm abilities feel fair?), builds an audience (people playing solo who want to compete with friends), and proves the concept (downloads and reviews) — all before a single line of networking code is written.
Key Insights
- The five-year Tenlach timeline is a case study in wrong sequencing. Every previous attempt started with multiplayer infrastructure (Nakama, Go servers, rollback netcode) before proving the core game was fun. This is building the stadium before confirming people want to watch the sport. Tenlach Arcade corrects the sequence: build the sport, prove it is worth watching, then build the stadium.
- The arcade/gauntlet mode is the MVP of game development. In startup terms: do not build the marketplace before proving people want the product. In game terms: do not build the multiplayer before proving the combat is fun. The minimum viable game is always single-player. Everything else is scaling.
- Many of the best multiplayer games launched as single-player first: Minecraft was single-player for its first public alpha. Fortnite launched as a single-player PvE game (Save the World) before Battle Royale existed. Even Among Us was released and largely ignored for two years before the multiplayer social dynamics caught fire during COVID. Single-player first is not a compromise — it is a proven strategy.
- If Tenlach Arcade ships and gets 1,000 players, multiplayer becomes urgent. If it ships and gets 12 players, the game design needs more work before multiplayer would help. The single-player launch is a litmus test for the concept itself. This is the diagnostic humility that was missing from every previous Tenlach attempt.
This post was composed through a conversation between Brett Owers and Claude Code (Anthropic). The content reflects Brett's recollection of each project and the lessons drawn from it. Some details may be approximate or omitted — the purpose is to paint an honest picture of a software engineer's development over time, not to serve as a precise historical record.