Quick/Comparisons
61 | Gaming | 7 |
62 | Workstation | 16 |
54 | AI/ML |
Compare with something else?
Compare GeForce RTX 4070 with another →8 |
46 | Energy Efficiency | 54 |
60 | Quick Comparison Final Score | 11 |
| AD104 | Graphics Processor | Polaris 21 |
| 7168 | Cores | 1024 |
| 224 / 80 | TMUs / ROPs | 64 / 16 |
| 6153.24 | Blender GPU | 137.81 |
| 131 | Forza Horizon 5 | 15 |
| 176 | The Witcher 3 | 15 |
| 212 | Counter-Strike 2 | 22 |
| 145 | Far Cry 6 | 19 |
| 106 | Hogwarts Legacy | 10 |
| 165 | Call of Duty | 20 |
| 100 | Ghost of Tsushima | 11 |
| 131 | Cyberpunk 2077 | 10 |
| 226 | Tomb Raider | 18 |
| 155 | Average FPS | 16 |
| 183 | 1080p High | 22 |
| 155 | 1080p Ultra | 16 |
| 120 | 1440p Ultra | 12 |
| 70 | 4K Ultra | - |
| Low | Margin of Error | Low |
| 206598 | GB6 Compute Score | 19139 |
| 359.5 img/sec | Background Blur | 34 img/sec |
| 210 img/sec | Face Detection | 17.6 img/sec |
| 7.61 Gpixels/sec | Horizon Detection | 0.69 Gpixels/sec |
| 10.4 Gpixels/sec | Edge Detection | 1.43 Gpixels/sec |
| 11.3 Gpixels/sec | Gaussian Blur | 1.15 Gpixels/sec |
| 1.97 Gpixels/sec | Feature Matching | 0.2 Gpixels/sec |
| 969.3 Gpixels/sec | Stereo Matching | 65.5 Gpixels/sec |
| 29061.4 FPS | Particle Physics | 2232.5 FPS |
| OpenCL | API | OpenCL |
| 29964 | G3D Mark Score | 3682 |
| 1193 | G2D Mark | 486 |
| 271 FPS | DirectX 11 | 25 FPS |
| 110 FPS | DirectX 12 | 21 FPS |
| 16985 Ops/s | GPU Compute | 1417 Ops/s |
| 30706 | GB6 ML Single Precision | 4551 |
| 43851 | GB6 ML Half Precision | 5280 |
| 23776 | GB6 ML Quantized | 3415 |
| 12146 | Image Classification (SP) | 1731 |
| 23932 | Image Segmentation (HP) | 2205 |
| 36321 | Image Super Resolution (Q) | 4705 |
| 52836 | Face Detection (HP) | 6423 |
| 179788 | Pose Estimation (Q) | 19379 |
| 3371 | Text Classification (SP) | 1172 |
| 5017 | Machine Translation (HP) | 1724 |
| 15164 | Object Detection (SP) | 1937 |
| 62500 | Depth Estimation (Q) | 8173 |
| 357489 | Style Transfer (SP) | 36502 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 198 GPixel/s | Pixel Fill Rate | 20 GPixel/s |
| 554 GTexel/s | Texture Fill Rate | 82 GTexel/s |
| 35.5 TFLOPS | FLOPS (FP32) | 2.6 TFLOPS |
| GDDR6X | Memory Type | GDDR5 |
| 12 GB | Memory Size | 4 GB |
| 1313 MHz | Memory Clock | 1750 MHz |
| 21000 Mbps | Effective Memory Speed | 7000 Mbps |
| 192-bit | Bus | 128-bit |
| No | ECC | No |
| 504.2 GB/s | Memory Bandwidth | 112 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 3.0 x8 |
| 220 W | TGP | 75 W |
| TSMC | Manufacturing | GlobalFoundries |
| 5 nm | Fabrication Process | 14 nm |
| 294 mm² | Die Size | 123 mm² |
| 35 billion | Transistor Count | 3 billion |
| 119.05 MTr/mm² | Transistor Density | 24.39 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.1 |
| 8.9 | CUDA | — |
| Yes | Ray Tracing | No |
| DLSS 3 | DLSS | No |
| 1.4a | DisplayPort | 1.4a |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| January 17, 2024 | Released | April 18, 2017 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Ada Lovelace | Architecture | GCN 4.0 |
| AD104 | GPU Codename | Polaris 21 |
| -Intel Core i7 14700Kor above | Recommended CPU | -Intel Core i5 8400or above |
| 21568 | Steel Nomad Lite Score | 1768 |
| 20882 | Time Spy | 1854 |
| 98166 | Solar Bay | — |
| 13122 | Port Royal | — |
| 50092 | Fire Strike | 6492 |
| 41630 | Wild Life Extreme | 3605 |
| 164777 | Night Raid | 25413 |
| 1980 MHz | Base Clock | 1175 MHz |
| 2475 MHz | Boost Clock | 1275 MHz |
| 7168 | Shading Units | 1024 |
| 224 | Texture Mapping Units (TMUs) | 64 |
| 80 | Render Output Units (ROPs) | 16 |
| 56 | Compute Units (Pipelines) | — |
| 224 | Tensor Cores | No |
| 56 | Ray-tracing Cores | No |
| 128KB per cluster | L1 Cache | 16KB per cluster |
| 48MB shared | L2 Cache | 1024KB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |