Network multiplayer optimization (delta synchronization, reconnect support, chat and UI improvements)

[MostCromulent]

Full Disclosure Up Front: I have zero coding ability. All code has been written by Claude Code AI based on natural language instruction. However that doesn't mean I have just accepted the first thing it spat out. I understand people may be skeptical about the "vibe coding" aspect of this, which is why I have attempted to verify the effectiveness of the code as much as I can through extensive automated testing, and why I have provided comprehensive documentation so there is full transparency for reviewers. If the devs have technical implementation questions or feedback I am going to have to rely on Claude to help me respond to that. I'm sorry if this is a mess; ultimately I am trying to help contribute to Forge in the best way I can. Yes, a comprehensive network play overhaul was probably overambitious as a first project by a non-coder.

Overview

This PR attempts to introduce major improvements to Forge's network play functionality.

These features and their implementation are comprehensively documented in Branch Documentation. In summary:

• Delta Synchronisation Protocol: The current network protocol transfers the entire game state to each player on every update. This is replaced with a new protocol which, after initialisation, only transfers changed properties in the game state on update. If there is a sync error defaults back to the current protocol to restore the full game state. Testing indicates this process massively reduces total bandwidth usage by around ~99.5% compared to status quo. (An average of 2.39 MB per game using new protocol vs 578.58 MB per game using current protocol across an automated 100-game test).

• Reconnection support: Allows disconnected players to reconnect to games with automatic state recovery. If a player cannot reconnect within a timeout period then they will be automatically handed to AI control so the game can continue. The host may manually skip the timeout and immediately hand control to AI.

• Enhanced Chat Notifications: Improved visibility for network play chat messages with better formatting and player join/leave notifications.

• Network UI Improvements: Better feedback during network operations including the prompt window identifying which player currently has priority and how long the game has been waiting for them to take an action; e.g. "Waiting for MostCromulent... (32 seconds)".

• Comprehensive Test Infrastructure: Automated headless network testing and analysis tools specifically developed to verify and debug the delta synchronisation protocol. Includes the ability to headlessly run large scale parallel AI vs AI games to completion using full network infrastructure, with comprehensive debug logging.

Dependencies

Dependencies between different features in the Branch are outlined in Feature Dependencies. In summary:

┌─────────────────────┐
│  Network Protocol   │─────────────────────────────► FOUNDATIONAL
│    (~4,500 lines)   │
└──────────┬──────────┘
           │
           ├──────────────────────────────────────────────────────┐
           │                                                      │
           ▼                                                      ▼
┌─────────────────────┐                              ┌─────────────────────┐
│ UI Improvements     │                              │ Reconnection Support│
│ (partial, ~200 ln)  │                              │    (~1,250 lines)   │
│ - Player ID sync    │                              │                     │
│ - Bandwidth metrics │                              │                     │
└─────────────────────┘                              └─────────────────────┘

┌─────────────────────┐
│  Chat Improvements  │─────────────────────────────► INDEPENDENT
│      (~200 lines)   │
└─────────────────────┘

┌─────────────────────┐
│   Testing Tools     │─────────────────────────────► INDEPENDENT
│    (~9,650 lines)   │
└─────────────────────┘

Testing

I have attempted to verify the soundness of the delta synchronisation protocol as thoroughly as I can alone using the automated testing infrastructure developed for this Branch. Details of testing capabilities are outlined at Testing Documentation.

Comprehensive Validation Results (29 January 2025) | Test log analysis and all 100 test logs available for review.

• A batch test of 100 AI vs AI games on full network infrastructure across a variety of player counts from 2-4 players, using over 200 unique decks to test a wide range of play conditions.
• 97% success rate with zero checksum errors. The three failed games were caused by timeouts due to complex board states, not the network protocol.
• Average bandwidth usage of 2.39 MB per game using new delta protocol vs 578.58 MB per game using current protocol.
• 99.6% average bandwidth savings.

Conclusion
Based on testing, the new delta synchronisation network protocol in this branch appears to solve a longstanding issue with Forge's multiplayer implementation, significantly reducing bandwidth usage and latency in network multiplayer games. The other features of this branch provide a range of quality-of-life improvements for network play.

I've taken this about as far as I can alone and would welcome any feedback and testing from others.

[Hanmac]

I will take a look at the serializeChangedOnly parts

i think some of the shuff could be added early

[tool4ever]

my developer brain tells me this should be cherry picked to the moon xD
on the other hand no one has worked on netplay in years, so maybe we could yolo it to some extent lol
I'll try and come up with a more nuanced opinion after first spending some time checking how it's looking for AI hallucinations

[MostCromulent]

I'll try and come up with a more nuanced opinion after first spending some time checking how it's looking for AI hallucinations

If it helps: when I first started I had the AI manually reading the network debug logs and reporting on the results. It was generally fine with identifying errors and bug fixing, but it wasn't good at mathematical calculations there were occasions when it would hallucinate the bandwidth numbers.

At that point I moved to having the log review and validation results strictly calculated through the log analysis tools rather than by the AI. The test results are directly generated by those tools.