Good question.
Sorry to disappoint but there is no observable effect on performance.
```c void GetPathPointsSmallHeavy() { GraphicsPath path; int x, y, length, width; PointF *dataPoints; double time_taken; for (int i = 0; i < 10; i++) { x = rand() % 1920; y = rand() % 1080; length = rand() / 1920; width = rand() / 1080;path.AddRectangle(Rect(x, y, length, width)); } dataPoints = new PointF[10 * 4]; auto started = std::chrono::high_resolution_clock::now(); for (int i = 0; i < 1000000; i++) { path.GetPathPoints(dataPoints, 10 * 4); } auto done =std::chrono::high_resolution_clock::now();std::cout<<std::chrono::duration_caststd::chrono::milliseconds(done - started).count()<< std::endl; delete [] dataPoints; } ```
The above snippet consistently yields 14 - 15ms on both master and this MR's branch.