IdeaWell

Project ideas from Hacker News discussions.

The unlikely story of Teardown Multiplayer

Original Article

Hacker News Discussion

📝 Discussion Summary (Click to expand)

3 Prevalent Themes in the Discussion

Theme Key Take‑aways & Representative Quotes
1. Deterministic multiplayer & networking challenges • The developers explain several replication strategies, ultimately favoring a “deterministic command stream” that the client can replay after the scene loads.
  > “Record the deterministic command stream, pass it to the joining client, and have that client apply all changes to the loaded scene before joining the game. The amount of data is much smaller … but applying the changes can take a while.” – sweepi
• The inconvenience of late‑join limitations is acknowledged.
  > “It’s only kind of inconvenient. By the time the scene gets to the point where join‑in‑progress is disabled it’s complete chaos anyway. Might as well restart the scene.” – florbo
2. Viability of cloud‑ or server‑side streaming as an alternative • Several users propose hosting the whole simulation on a single powerful machine and streaming video to players, noting the trade‑off between perfect consistency and bandwidth/latency constraints.
  > “Wouldn’t it have been simpler … to host the game on a single machine and just stream each player’s camera?” – DimitriBouriez
  > “Video streams are not known for their low bandwidth needs, let alone adding in RTT latency for inputs.” – dwroberts
3. Confidence in the developers & community optimism • Long‑time fans praise the engineering behind deterministic multiplayer and the way Coffee Stain handles its titles, suggesting a bright future for the game.
  > “I enjoyed playing Teardown when it first came out. It was already a technological marvel … Clearly top tier engineers.” – kajkojednojajko
  > “It seems like Coffee Stain generally handles the games they publish pretty well. They also have Valheim and Deep Rock Galactic published under them, both of which have been reasonably successful and long running games.” – VoidWhisperer

These three themes capture the dominant topics of conversation: the technical intricacies of building a deterministic multiplayer experience, debates over alternative architecture choices like cloud streaming, and the overall optimism that the developers will deliver a solid multiplayer implementation.

🚀 Project Ideas

Teardown JoinSync Engine

Summary

  • Eliminates the need for large scene file transfers by streaming a compact deterministic command log, allowing players to join sessions instantly even on low‑bandwidth connections.
  • Core value: seamless multiplayer join‑in‑progress with near‑zero latency and minimal bandwidth.

Details

Key Value
Target Audience Teardown modders, voxel‑based multiplayer developers, indie studios building deterministic multiplayer games
Core Feature Deterministic command replay system with adaptive buffering and automatic state reconciliation for mid‑session joins
Tech Stack C++ backend, ECS physics, QUIC/UDP transport, Node.js admin API, WebAssembly UI
Difficulty Medium
Monetization Revenue-ready: Subscription $5/mo per 100 concurrent players

Notes

  • HN commenters repeatedly lamented that “option 3” required replaying the deterministic command stream and that “we still limit the buffer size…” – this solves that bottleneck.
  • Low‑end devices like the Steam Deck could join sessions without stalling, addressing worries about the game becoming too heavy for portable play.

CloudPhysics Hosting for Deterministic Multiplayer Games

Summary

  • Provides a SaaS that runs the entire deterministic physics simulation on powerful servers and streams only compressed state updates to clients, removing the need for each player to host or for clients to exchange massive voxel data.
  • Core value: Consistent, low‑latency multiplayer across all devices, with negligible install size and no piracy concerns.

Details

Key Value
Target Audience Indie developers of physics‑heavy multiplayer games, community servers, esports organizers
Core Feature Centralized deterministic simulation service that streams minimal state packets and accepts client inputs
Tech Stack Go microservices, QUIC, WebRTC data channels, Rust compute kernels, PostgreSQL for session state
Difficulty High
Monetization Revenue-ready: Pay‑per‑hour $0.02 per compute-second

Notes

  • Commenters asked “Wouldn't it have been simpler (even if technically heavy) to host the game on a single machine and just stream each player's camera?” – this service does exactly that but with deterministic sync instead of video.
  • Eliminates the “average answer to be zero” problem of finding a host, making it easy for anyone to play together regardless of hardware.

DeterministicPhysicsKit

Summary

  • An open‑source, cross‑platform physics engine library that guarantees bit‑identical simulation results on x86, ARM, and consoles by using deterministic floating‑point math and fixed‑point fallback, easing multiplayer network code for voxel games.
  • Core value: Guarantees deterministic outcomes across disparate hardware, removing a major source of desync bugs.

Details

Key Value
Target Audience Game developers targeting multiplatform releases, especially those using voxel or block‑based physics
Core Feature Deterministic math kernels (fixed‑point, IEEE‑754 safe) with optional SIMD acceleration and built‑in networking sync primitives
Tech Stack Rust core, SIMD, WASM for browser, Python bindings, CI for cross‑arch testing
Difficulty Medium
Monetization Revenue-ready: Enterprise license $1,500 one‑time + support subscription

Notes

  • Discussions about “floating point determinism” and the need to “restrict yourself to using only the basic arithmetic operations” are solved out‑of‑the‑box by this kit.
  • Would have been loved by commenters who wanted a reliable way to sync without worrying about CPU architecture differences.

Read Later