Nvidia GeForce RTX 4070 SUPER vs Apple M5 Pro GPU (20-core)
63
out 100
VS
43
out 100
Compared 2 processors: Nvidia GeForce RTX 4070 SUPER, vs Apple M5 Pro GPU (20-core). Below are specs, tests, pros and cons for each device
Review
Games
Performance in DirectX, OpenCL, Vulkan
AI/ML
AI workload testing
Professional Apps
3D modeling, video editing, rendering
Energy Efficiency
Power usage in various scenarios
Tests & specs
Specs & test comparisonGames
| Game title | 1080p (High) 1080p (Ultra) 1440p (Ultra) 4K (Ultra) | |
 Forza Horizon 5 | 184 128 116 89 | n/a n/a n/a n/a |
 The Witcher 3 | 204 175 127 68 | n/a n/a n/a n/a |
 Counter-Strike 2 | 283 214 156 83 | n/a n/a n/a n/a |
 Far Cry 6 | 157 142 127 76 | n/a n/a n/a n/a |
 Hogwarts Legacy | 132 110 87 48 | n/a n/a n/a n/a |
 COD Modern Warfare 3 | 171 165 128 83 | n/a n/a n/a n/a |
 Ghost of Tsushima | 118 99 85 50 | n/a n/a n/a n/a |
 Cyberpunk 2077 | 154 136 92 40 | n/a n/a n/a n/a |
 Shadow of the Tomb Raider | 242 224 176 93 | n/a n/a n/a n/a |
| 1080p High | 183 | n/a |
| 1080p Ultra | 155 | n/a |
| 1440p Ultra | 122 | n/a |
| 4K Ultra | 70 | n/a |
| Deviation | Low | n/a |
Benchmarks
| Time Spy | 20901 | n/a |
| Solar Bay | 97900 | 46898 |
| Port Royal | 13112 | n/a |
| Fire Strike | 50143 | n/a |
| Night Raid | 164308 | n/a |
| Background Blur | 359.5 frames/s | 177.6 frames/s |
| Face Detection | 210 frames/s | 114.5 frames/s |
| Horizon Detection | 7.61 Gpix/sec | 3.19 Gpix/sec |
| Edge Detection | 10.4 Gpix/sec | 4.45 Gpix/sec |
| Gaussian Blur | 11.3 Gpix/sec | 4.59 Gpix/sec |
| Feature Matching | 1.97 Gpix/sec | 0.96 Gpix/sec |
| Stereo Matching | 969.3 Gpix/sec | 250.4 Gpix/sec |
| Particle Physics | 29061.4 fps | 12529.9 fps |
| API | OpenCL | OpenCL |
| G2D Mark | 1178 | n/a |
| DirectX 11 | 274 fps | n/a |
| DirectX 12 | 108 fps | n/a |
| GPU Compute | 17275 Ops/s | n/a |
| Aztec Ruins (Normal) | 1132.4 fps | n/a |
| Car Chase | 981.8 fps | n/a |
| Manhattan 3.1.1 (1440p) | 840.6 fps | n/a |
| T-Rex | 4325.3 fps | n/a |
| ALU 2 | 6890.2 fps | n/a |
| Driver Overhead 2 | 290.1 fps | n/a |
| API | OpenGL | n/a |
| OS | Windows | n/a |
| Tessellation | 7090.8 | n/a |
| Image Classification (SP) | n/a | 8870 |
| Image Segmentation (HP) | n/a | 28795 |
| Horizon Detection | n/a | 34820 |
| Face Detection (HP) | n/a | 55699 |
| Pose Estimation (Q) | n/a | 212347 |
| Text Classification (SP) | n/a | 3018 |
| Machine Translation (HP) | n/a | 9102 |
| Object Detection (SP) | n/a | 8163 |
| Depth Estimation (Q) | n/a | 70908 |
| Style Transfer (SP) | n/a | 152297 |
| Framework | n/a | Core ML |
| Backend | n/a | GPU |
Tech Specs
Specs & test comparisonGeneral Info
| Manufacturer | Nvidia | Apple |
| Build | Discrete | Integrated |
| Announced | January, 17, 2024 | March, 3, 2026 |
| Form Factor | PC use | Laptop use |
| Use case | Gaming | Professional |
| Segment | Mid-range | Mid-range |
| Architecture | Ada Lovelace | Apple M GPU |
| Codename | AD104 | Custom |
| Price | 599 | n/a |
GPU
| GPU Frequency | 1980 MHz | n/a |
| Boost GPU Frequency | 2475 MHz | 1620 MHz |
| Shader Units | 7168 | 2560 |
| TMUs | 224 | 160 |
| ROPs | 80 | 80 |
| Execution Units | 56 | 320 |
| Ray Tracing Cores | 56 | n/a |
| Tensor Cores | 224 | n/a |
| L1 cache | 128 KB/core | n/a |
| L2 cache | 48000 KB/core | n/a |
| Instructions per Clock | 2 IPC | 2 IPC |
Performance
| Pixel Fill Rate | 198 Gpix/sec | 130 Gpix/sec |
| Texture Fill Rate | 554 GTex/s | 259 GTex/s |
| FLOPS | 35.5 TFLOPS | 8.3 TFLOPS |
Process
| Interface | PCIe 4.0 x16 | Custom |
| TGP | 220 W | 28 W |
| Process | TSMC | TSMC |
| Process tech. | 5 nm | 3 nm |
| Max temperature | n/a | 100 °C |
| Transistor Size | 294 mm² | n/a |
| Transistor count | 35 billion | 28 billion |
| Transistor density | 119.05 MTr/mm² | n/a |
Memory
| Interface | GDDR6X | System Shared |
| Memory Frequency | 1313 MHz | 9600 MHz |
| Bus | 192 bit | 256 bit |
| ECC support | No | No |
| Effective Memory Speed | 21000 MB/s | n/a |
| Memory Size | 12 GB | n/a |
Max Throughput
API
| DirectX | 12 | n/a |
| Vulkan | 1.3 | n/a |
| OpenGL | 4.6 | n/a |
| CUDA | 8.9 | n/a |
| DisplayPort | 1.4a | n/a |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 3 | No |
Results
After analyzing the data, we consider Nvidia GeForce RTX 4070 SUPER the clear winner in this comparison.
Survey
Total Votes: 0
