FramePro Reader is a console application that reads FramePro (.framepro) profiling files and generates comprehensive performance analysis reports with a focus on multithreading optimization.

• Reads FramePro save files (.framepro format)
• Analyzes session metadata (frame count, timer frequency, thread information)
• Identifies hot functions with detailed performance metrics
• Groups functions by threads for multithreading analysis
• Generates frame-by-frame performance data
• Exports data in structured JSON format

dotnet run --project FrameProReader "path/to/your/file.framepro"

• TotalTimeMs: Total execution time across all frames
• TotalCount: Total number of function calls
• AvgTimePerFrameMs: Average execution time per frame
• MaxTimePerFrameMs: Maximum execution time in any single frame
• AvgCountPerFrame: Average number of calls per frame
• MaxCountPerFrame: Maximum number of calls in any single frame

• ThreadId: Unique identifier for the thread
• ThreadName: Human-readable thread name
• ThreadUtilizationPercent: Percentage of thread utilization
• IsMainThread: Whether function runs on main thread (UI blocking)
• IsRenderThread: Whether function runs on render thread (FPS impact)
• IsWorkerThread: Whether function runs on worker thread (parallelizable)

The JSON files provide data to identify:

• Functions with high TotalTimeMs (optimization priority)
• Functions with high MaxTimePerFrameMs (cause frame drops)
• Functions with high ThreadUtilizationPercent (thread saturation)

• Functions on main thread (IsMainThread: true) blocking UI
• Functions on render thread (IsRenderThread: true) affecting FPS
• Overloaded threads (ThreadUtilizationPercent > 80%)

• Functions suitable for worker threads (IsWorkerThread: true)
• Functions with high call frequency (AvgCountPerFrame)
• Functions with inconsistent performance (high MaxTimePerFrameMs vs AvgTimePerFrameMs)

• Framework: .NET 8.0
• Dependencies: FrameProCore.dll, SCLCoreCLR.dll
• Output Format: Structured JSON using System.Text.Json
• Performance: Handles large datasets efficiently (212 functions ×ばつ 254 frames)

1. Run the tool on your .framepro file
2. Open functions_analysis.json to identify top performance issues
3. Filter by IsMainThread: true to find UI-blocking functions
4. Sort by TotalTimeMs to prioritize optimization efforts
5. Use frame_analysis.json to analyze specific problematic frames
6. Cross-reference thread utilization to balance workload

{
 "FunctionName": "Event Wait",
 "ThreadId": 5032,
 "ThreadName": "TaskGraph Render Thread 2",
 "TotalTimeMs": 37150.96,
 "TotalCount": 6426,
 "MaxTimePerFrameMs": 519.17,
 "MaxCountPerFrame": 38,
 "AvgTimePerFrameMs": 146.84,
 "AvgCountPerFrame": 25.40,
 "ThreadUtilizationPercent": 100,
 "IsMainThread": false,
 "IsRenderThread": true,
 "IsWorkerThread": true,
 "ThreadPriority": 0
}