Geometric Mean Speedup
46.2x
Across 27 comparative benchmarks
Pure Compute Speedup
34.5x
10 numerical benchmarks
Engine Core Speedup
54.8x
17 engine API benchmarks
Comparison Groups
9
Grouped domain subsystems
Peak Speedup
3328.0x
AStar2D
Side-by-side grouped evaluations measuring specific domain subsystems against baselines across languages and paradigms.
| Language / Target |
Implementation |
Latency (ms) |
Speedup vs GDScript |
Ratio vs Crystal |
| Crystal |
Native LibGodot / LLVM O3 |
43.01 ms |
28.0x faster |
1.0x (Baseline) |
| C++ |
GCC / Clang -O3 |
27.36 ms |
44.0x faster |
1.57x faster |
| Rust |
rustc -O |
28.01 ms |
43.0x faster |
1.54x faster |
| C# |
.NET 8 RyuJIT Release |
49.26 ms |
24.4x faster |
1.15x slower |
| GDScript |
Godot VM Bytecode (Headless) |
1203.67 ms |
1.0x (Baseline) |
28.0x slower than Crystal |
| Benchmark |
Description |
Crystal |
GDScript |
Speedup |
| CellularNoise |
FastNoiseLite 2D Cellular Voronoi noise across 500x500 grid (250k samples) |
0.39 ms |
57.94 ms |
148.6x faster |
| PerlinNoise |
FastNoiseLite 2D Perlin noise across 500x500 grid (250k samples) |
0.42 ms |
35.43 ms |
84.4x faster |
| SimplexNoise |
FastNoiseLite 3D Simplex Smooth noise across 100k samples |
0.19 ms |
20.71 ms |
109.0x faster |
| Benchmark |
Description |
Crystal |
GDScript |
Speedup |
| CompileTimes |
Compilation latency and output binary size across Crystal, C++, C#, Rust (Debug vs Release) |
9303.08 ms |
- |
Native |
| Language / Target |
Implementation |
Latency (ms) |
Speedup vs GDScript |
Ratio vs Crystal |
| Crystal |
Native LibGodot / LLVM O3 |
18.1 ms |
67.2x faster |
1.0x (Baseline) |
| C++ |
GCC / Clang -O3 |
18.34 ms |
66.3x faster |
1.01x slower |
| Rust |
rustc -O |
18.79 ms |
64.7x faster |
1.04x slower |
| C# |
.NET 8 RyuJIT Release |
41.39 ms |
29.4x faster |
2.27x slower |
| GDScript |
Godot VM Bytecode (Headless) |
1216.58 ms |
1.0x (Baseline) |
67.2x slower than Crystal |
| Language / Target |
Implementation |
Latency (ms) |
Speedup vs GDScript |
Ratio vs Crystal |
| Crystal |
Native LibGodot / LLVM O3 |
27.5 ms |
81.4x faster |
1.0x (Baseline) |
| C++ |
GCC / Clang -O3 |
24.52 ms |
91.3x faster |
1.12x faster |
| Rust |
rustc -O |
27.07 ms |
82.7x faster |
1.02x faster |
| C# |
.NET 8 RyuJIT Release |
32.15 ms |
69.6x faster |
1.16x slower |
| GDScript |
Godot VM Bytecode (Headless) |
2238.78 ms |
1.0x (Baseline) |
81.4x slower than Crystal |
| Language / Target |
Implementation |
Latency (ms) |
Speedup vs GDScript |
Ratio vs Crystal |
| Crystal |
Native LibGodot / LLVM O3 |
3.94 ms |
142.8x faster |
1.0x (Baseline) |
| C++ |
GCC / Clang -O3 |
3.83 ms |
146.9x faster |
1.03x faster |
| Rust |
rustc -O |
3.99 ms |
141.0x faster |
1.01x slower |
| C# |
.NET 8 RyuJIT Release |
13.77 ms |
40.9x faster |
3.45x slower |
| GDScript |
Godot VM Bytecode (Headless) |
562.55 ms |
1.0x (Baseline) |
142.8x slower than Crystal |
| Benchmark |
Description |
Crystal |
GDScript |
Speedup |
| NodeGroups |
20,000 nodes partitioned into 10 groups, group assignment & queries |
1.44 ms |
11.98 ms |
8.3x faster |
| NodeLifecycle |
Node2D allocation, property mutation, add_child, remove_child, free (20k) |
4.09 ms |
25.72 ms |
6.3x faster |
| TreeTraversal |
Scene tree recursive traversal & property inspection (20k nodes) |
0.0 ms |
8.35 ms |
1.0x faster |
| Language / Target |
Implementation |
Latency (ms) |
Speedup vs GDScript |
Ratio vs Crystal |
| Crystal |
Native LibGodot / LLVM O3 |
18.03 ms |
29.3x faster |
1.0x (Baseline) |
| C++ |
GCC / Clang -O3 |
16.58 ms |
31.9x faster |
1.09x faster |
| Rust |
rustc -O |
15.54 ms |
34.0x faster |
1.16x faster |
| C# |
.NET 8 RyuJIT Release |
42.76 ms |
12.4x faster |
2.38x slower |
| GDScript |
Godot VM Bytecode (Headless) |
528.86 ms |
1.0x (Baseline) |
29.3x slower than Crystal |
| Benchmark |
Description |
Crystal |
GDScript |
Speedup |
| ShaderCompilation |
Shader compilation throughput across 1,000 permutations |
0.44 ms |
92.49 ms |
210.2x faster |
| ShaderUniforms |
High-frequency uniform parameter dispatch (50k updates) |
0.21 ms |
13.66 ms |
65.0x faster |
| VisualShader |
VisualShader node graph creation and connection (200 graphs) |
0.02 ms |
4.79 ms |
239.5x faster |