Quick/Comparisons
41 | Gaming | 51 |
40 | Workstation | 56 |
38 | AI/ML |
Compare with something else?
Compare GeForce RTX 4070 with another →33 |
56 | Energy Efficiency | 41 |
43 | Quick Comparison Final Score | 49 |
| AD106 (GN21-X6) | Graphics Processor | GA103 |
| 4608 | Cores | 7424 |
| 144 / 48 | TMUs / ROPs | 232 / 96 |
| 3587.08 | Blender GPU | 2925.42 |
| 90 | Forza Horizon 5 | — |
| 132 | The Witcher 3 | — |
| 178 | Counter-Strike 2 | — |
| 112 | Far Cry 6 | — |
| 75 | Hogwarts Legacy | — |
| 104 | Call of Duty | — |
| 68 | Ghost of Tsushima | — |
| 87 | Cyberpunk 2077 | — |
| 139 | Tomb Raider | — |
| 109 | Average FPS | — |
| 131 | 1080p High | — |
| 109 | 1080p Ultra | — |
| 81 | 1440p Ultra | — |
| 46 | 4K Ultra | — |
| Low | Margin of Error | — |
| 115081 | GB6 Compute Score | 144566 |
| 158.5 img/sec | Background Blur | 222.1 img/sec |
| 116.5 img/sec | Face Detection | 132.2 img/sec |
| 4.35 Gpixels/sec | Horizon Detection | 6.83 Gpixels/sec |
| 5.67 Gpixels/sec | Edge Detection | 9.64 Gpixels/sec |
| 6.44 Gpixels/sec | Gaussian Blur | 7.81 Gpixels/sec |
| 1.36 Gpixels/sec | Feature Matching | 1.27 Gpixels/sec |
| 528.5 Gpixels/sec | Stereo Matching | 615.9 Gpixels/sec |
| 16372.3 FPS | Particle Physics | 18134.8 FPS |
| OpenCL | API | OpenCL |
| 19527 | G3D Mark Score | 17167 |
| 767 | G2D Mark | 713 |
| 178 FPS | DirectX 11 | 132 FPS |
| 85 FPS | DirectX 12 | 84 FPS |
| 8389 Ops/s | GPU Compute | 7423 Ops/s |
| 21186 | GB6 ML Single Precision | 15704 |
| 34404 | GB6 ML Half Precision | 26550 |
| 16559 | GB6 ML Quantized | 9885 |
| 8642 | Image Classification (SP) | 6982 |
| 20635 | Image Segmentation (HP) | 15050 |
| 26378 | Image Super Resolution (Q) | 14412 |
| 39776 | Face Detection (HP) | 27475 |
| 105983 | Pose Estimation (Q) | 66491 |
| 2995 | Text Classification (SP) | 2852 |
| 4507 | Machine Translation (HP) | 3306 |
| 10198 | Object Detection (SP) | 8121 |
| 41078 | Depth Estimation (Q) | 24442 |
| 190315 | Style Transfer (SP) | 156421 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 59-104 GPixel/s | Pixel Fill Rate | 101-160 GPixel/s |
| 177-313 GTexel/s | Texture Fill Rate | 244-386 GTexel/s |
| 11.3-20 TFLOPS | FLOPS (FP32) | 15.6-24.7 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 8 GB | Memory Size | 16 GB |
| 2000 MHz | Memory Clock | 2000 MHz |
| 16000 Mbps | Effective Memory Speed | 16000 Mbps |
| 128-bit | Bus | 256-bit |
| No | ECC | Yes |
| 256 GB/s | Memory Bandwidth | 512 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 35-140 W (configurable) | TGP | 80-165 W (configurable) |
| TSMC | Manufacturing | Samsung |
| 5 nm | Fabrication Process | 8 nm |
| 186 mm² | Die Size | 496 mm² |
| 22.9 billion | Transistor Count | 22 billion |
| 123.12 MTr/mm² | Transistor Density | 44.35 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.9 | CUDA | 8.6 |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | DLSS 2 |
| 1.4a | DisplayPort | 1.4 |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| February 22, 2023 | Released | March 30, 2022 |
| Laptop | Case | Laptop |
| Gaming | Purpose | Professional |
| High-end | Segment | High-end |
| Ada Lovelace | Architecture | Ampere |
| AD106 (GN21-X6) | GPU Codename | GA103 |
| -Intel Core Ultra 7 155Hor above | Recommended CPU | -Intel Core i9 12900HXor above |
| 12729 | Steel Nomad Lite Score | — |
| 12518 | Time Spy | 8880 |
| 54826 | Solar Bay | 32222 |
| 7403 | Port Royal | 5730 |
| 30044 | Fire Strike | 24599 |
| 23244 | Wild Life Extreme | 16879 |
| 113720 | Night Raid | 27155 |
| 735-2070 MHz (configurable) | Base Clock | — |
| 1230-2175 MHz (configurable) | Boost Clock | 1050-1663 MHz (configurable) |
| 4608 | Shading Units | 7424 |
| 144 | Texture Mapping Units (TMUs) | 232 |
| 48 | Render Output Units (ROPs) | 96 |
| 36 | Compute Units (Pipelines) | 58 |
| 144 | Tensor Cores | 232 |
| 36 | Ray-tracing Cores | 58 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 32MB shared | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |