{
  "schema": "sync-camera-research-findings-v1",
  "researchDates": [
    "2026-09-21"
  ],
  "findings": [
    {
      "cycle": "4D",
      "title": "Packaged Electron Native Shared Memory Ring Buffer Bypass",
      "finding": "Bypassing Chromium's CoreMedia/AVFoundation camera capture pipeline entirely in packaged Electron by attaching directly to the daemon's POSIX shared memory ring buffer (/tmp/SyncCamera.frames) via IPC and preallocated buffer pools completely avoids CMIO DAL IPC, macOS CoreMedia frame drops, and browser IPC marshalling. Frame intake executes with zero heap allocations per frame (0 bytes/s vs. ~497 MB/s in unpooled Buffers), and delivers raw BGRA pixels directly to WebGL2 textures (gl.texImage2D) in < 0.5 ms. Unique delivered frame rate achieves 59.90+ FPS with 0 low-rate intervals (<50 FPS) and p99 delivery latency under 20 ms."
    }
  ]
}