Quick/Comparisons
61 | Gaming | 3 |
62 | Workstation | 11 |
54 | AI/ML |
Compare with something else?
Compare GeForce RTX 4070 with another →| — |
46 | Energy Efficiency | 71 |
60 | Quick Comparison Final Score | 18 |
| AD104 | Graphics Processor | Coffee Lake GT2 |
| 7168 | Cores | 192 |
| 224 / 80 | TMUs / ROPs | 24 / 3 |
| 6153.24 | Blender GPU | 74.94 |
| 131 | Forza Horizon 5 | — |
| 176 | The Witcher 3 | — |
| 212 | Counter-Strike 2 | — |
| 145 | Far Cry 6 | — |
| 106 | Hogwarts Legacy | — |
| 165 | Call of Duty | — |
| 100 | Ghost of Tsushima | — |
| 131 | Cyberpunk 2077 | — |
| 226 | Tomb Raider | — |
| 155 | Average FPS | — |
| 183 | 1080p High | — |
| 155 | 1080p Ultra | — |
| 120 | 1440p Ultra | — |
| 70 | 4K Ultra | — |
| Low | Margin of Error | — |
| 206598 | GB6 Compute Score | 5810 |
| 359.5 img/sec | Background Blur | 11.2 img/sec |
| 210 img/sec | Face Detection | 6.77 img/sec |
| 7.61 Gpixels/sec | Horizon Detection | 0.21 Gpixels/sec |
| 10.4 Gpixels/sec | Edge Detection | 0.31 Gpixels/sec |
| 11.3 Gpixels/sec | Gaussian Blur | 0.22 Gpixels/sec |
| 1.97 Gpixels/sec | Feature Matching | 0.08 Gpixels/sec |
| 969.3 Gpixels/sec | Stereo Matching | 21.4 Gpixels/sec |
| 29061.4 FPS | Particle Physics | 794.3 FPS |
| OpenCL | API | OpenCL |
| 29964 | G3D Mark Score | 1227 |
| 1193 | G2D Mark | 294 |
| 271 FPS | DirectX 11 | 10 FPS |
| 110 FPS | DirectX 12 | 5 FPS |
| 16985 Ops/s | GPU Compute | 496 Ops/s |
| 30706 | GB6 ML Single Precision | — |
| 43851 | GB6 ML Half Precision | — |
| 23776 | GB6 ML Quantized | — |
| 12146 | Image Classification (SP) | — |
| 23932 | Image Segmentation (HP) | — |
| 36321 | Image Super Resolution (Q) | — |
| 52836 | Face Detection (HP) | — |
| 179788 | Pose Estimation (Q) | — |
| 3371 | Text Classification (SP) | — |
| 5017 | Machine Translation (HP) | — |
| 15164 | Object Detection (SP) | — |
| 62500 | Depth Estimation (Q) | — |
| 357489 | Style Transfer (SP) | — |
| ONNX | Framework | — |
| DirectML | Backend | — |
| 198 GPixel/s | Pixel Fill Rate | 3-4 GPixel/s |
| 554 GTexel/s | Texture Fill Rate | 24-30 GTexel/s |
| 35.5 TFLOPS | FLOPS (FP32) | 0.4-0.5 TFLOPS |
| GDDR6X | Memory Type | System Shared |
| 12 GB | Memory Size | System RAM Dependent |
| 1313 MHz | Memory Clock | System RAM Dependent |
| 21000 Mbps | Effective Memory Speed | System RAM Dependent |
| 192-bit | Bus | — |
| No | ECC | No |
| 504.2 GB/s | Memory Bandwidth | — |
| PCIe 4.0 x16 | Interface | Ring Bus |
| 220 W | TGP | 15 W |
| TSMC | Manufacturing | Intel |
| 5 nm | Fabrication Process | 14 nm |
| 294 mm² | Die Size | 49 mm² |
| 35 billion | Transistor Count | 0.19 billion |
| 119.05 MTr/mm² | Transistor Density | 3.88 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.1.103 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.1 |
| 8.9 | CUDA | — |
| Yes | Ray Tracing | No |
| DLSS 3 | DLSS | No |
| 1.4a | DisplayPort | 1.2 |
| Nvidia | Vendor | Intel |
| Discrete | Build | Integrated |
| January 17, 2024 | Released | April 3, 2018 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Generic |
| Mid-range | Segment | Entry-level |
| Ada Lovelace | Architecture | Generation 9.5 |
| AD104 | GPU Codename | Coffee Lake GT2 |
| -Intel Core i7 14700Kor above | Recommended CPU | -Intel Core i5 8300Hor above |
| 21568 | Steel Nomad Lite Score | 338 |
| 20882 | Time Spy | 438 |
| 98166 | Solar Bay | — |
| 13122 | Port Royal | — |
| 50092 | Fire Strike | 1328 |
| 41630 | Wild Life Extreme | 762 |
| 164777 | Night Raid | 5037 |
| 1980 MHz | Base Clock | 350 MHz |
| 2475 MHz | Boost Clock | 1000-1250 MHz (configurable) |
| 7168 | Shading Units | 192 |
| 224 | Texture Mapping Units (TMUs) | 24 |
| 80 | Render Output Units (ROPs) | 3 |
| 56 | Compute Units (Pipelines) | 24 |
| 224 | Tensor Cores | — |
| 56 | Ray-tracing Cores | — |
| 128KB per cluster | L1 Cache | — |
| 48MB shared | L2 Cache | — |
| 2 IPC | Instructions Per Cycle | 2 IPC |