Nvidia GeForce RTX 3070 vs Apple M4 GPU (10-Core)
45
out 100
VS
33
out 100
Compared 2 processors: Nvidia GeForce RTX 3070, vs Apple M4 GPU (10-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 | 151 98 85 64 | n/a n/a n/a n/a |
 The Witcher 3 | 171 141 104 55 | n/a n/a n/a n/a |
 Counter-Strike 2 | 241 195 134 78 | n/a n/a n/a n/a |
 Far Cry 6 | 131 119 95 56 | n/a n/a n/a n/a |
 Hogwarts Legacy | 97 81 60 33 | n/a n/a n/a n/a |
 COD Modern Warfare 3 | 120 116 81 53 | n/a n/a n/a n/a |
 Ghost of Tsushima | 88 74 57 31 | n/a n/a n/a n/a |
 Cyberpunk 2077 | 105 93 60 26 | n/a n/a n/a n/a |
 Shadow of the Tomb Raider | 164 151 112 59 | n/a n/a n/a n/a |
| 1080p High | 141 | n/a |
| 1080p Ultra | 119 | n/a |
| 1440p Ultra | 88 | n/a |
| 4K Ultra | 51 | n/a |
| Deviation | Low | n/a |
Benchmarks
| Time Spy | 13641 | n/a |
| Solar Bay | 64184 | n/a |
| Port Royal | 8309 | n/a |
| Fire Strike | 34034 | n/a |
| Night Raid | 120956 | n/a |
| Background Blur | 161.6 frames/s | 72.2 frames/s |
| Face Detection | 117.2 frames/s | 48.3 frames/s |
| Horizon Detection | 5.89 Gpix/sec | 1.4 Gpix/sec |
| Edge Detection | 8.05 Gpix/sec | 1.94 Gpix/sec |
| Gaussian Blur | 6.65 Gpix/sec | 1.5 Gpix/sec |
| Feature Matching | 1.25 Gpix/sec | 0.53 Gpix/sec |
| Stereo Matching | 550.6 Gpix/sec | 123.3 Gpix/sec |
| Particle Physics | 17613.1 fps | 4938.4 fps |
| API | OpenCL | OpenCL |
| G2D Mark | 998 | n/a |
| DirectX 11 | 181 fps | n/a |
| DirectX 12 | 85 fps | n/a |
| GPU Compute | 11282 Ops/s | n/a |
| Aztec Ruins (Normal) | 856 fps | 344.5 fps |
| Car Chase | 668 fps | 259.9 fps |
| Manhattan 3.1.1 (1440p) | 592.9 fps | 167.5 fps |
| T-Rex | 3078.4 fps | 914.7 fps |
| ALU 2 | 4531.5 fps | 432.2 fps |
| Driver Overhead 2 | 283 fps | 648.6 fps |
| API | OpenGL | Metal |
| OS | Windows | OS X |
| Tessellation | 4563.3 | n/a |
Tech Specs
Specs & test comparisonGeneral Info
| Manufacturer | Nvidia | Apple |
| Build | Discrete | Integrated |
| Announced | October, 22, 2020 | May, 7, 2024 |
| Form Factor | PC use | Laptop use |
| Use case | Gaming | Professional |
| Segment | Mid-range | Mid-range |
| Architecture | Ampere | Apple M GPU |
| Codename | GA104 | Custom |
| Price | 499 | n/a |
GPU
| GPU Frequency | 1500 MHz | 500 MHz |
| Boost GPU Frequency | 1725 MHz | 1800 MHz |
| Shader Units | 5888 | 1280 |
| TMUs | 184 | 80 |
| ROPs | 96 | 40 |
| Execution Units | 46 | 160 |
| Ray Tracing Cores | 46 | n/a |
| Tensor Cores | 184 | n/a |
| L1 cache | 128 KB/core | n/a |
| L2 cache | 4000 KB/core | n/a |
| Instructions per Clock | 2 IPC | 2 IPC |
Performance
| Pixel Fill Rate | 166 Gpix/sec | 72 Gpix/sec |
| Texture Fill Rate | 317 GTex/s | 144 GTex/s |
| FLOPS | 20.3 TFLOPS | 4.6 TFLOPS |
Process
| Interface | PCIe 4.0 x16 | Custom |
| TGP | 220 W | 18 W |
| Process | Samsung | TSMC |
| Process tech. | 8 nm | 3 nm |
| Max temperature | n/a | 100 °C |
| Transistor Size | 392 mm² | n/a |
| Transistor count | 17 billion | 28 billion |
| Transistor density | 43.37 MTr/mm² | n/a |
Memory
| Interface | GDDR6 | System Shared |
| Memory Frequency | 1750 MHz | 7500 MHz |
| Bus | 256 bit | 128 bit |
| ECC support | No | No |
| Effective Memory Speed | 14000 MB/s | 15000 MB/s |
| Memory Size | 8 GB | n/a |
Max Throughput
API
| DirectX | 12 | n/a |
| Vulkan | 1.3 | n/a |
| OpenGL | 4.6 | n/a |
| CUDA | 8.6 | n/a |
| DisplayPort | 1.4a | n/a |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 2 | No |
Results
After analyzing the data, we consider Nvidia GeForce RTX 3070 the clear winner in this comparison.
Survey
Total Votes: 0
