Quick/Comparisons
43 | Gaming | 31 |
44 | Workstation | 38 |
35 | AI/ML |
Compare with something else?
Compare GeForce RTX 3070 with another →27 |
35 | Energy Efficiency | 43 |
42 | Quick Comparison Final Score | 33 |
| GA104 | Graphics Processor | DG2-512 |
| 5888 | Cores | 3584 |
| 184 / 96 | TMUs / ROPs | 224 / 112 |
| 3368.42 | Blender GPU | 2033.41 |
| 99 | Forza Horizon 5 | 68 |
| 143 | The Witcher 3 | 114 |
| 194 | Counter-Strike 2 | 108 |
| 121 | Far Cry 6 | 87 |
| 80 | Hogwarts Legacy | 50 |
| 113 | Call of Duty | 69 |
| 74 | Ghost of Tsushima | 51 |
| 94 | Cyberpunk 2077 | 69 |
| 153 | Tomb Raider | 98 |
| 119 | Average FPS | 79 |
| 141 | 1080p High | 98 |
| 119 | 1080p Ultra | 79 |
| 88 | 1440p Ultra | 59 |
| 50 | 4K Ultra | 33 |
| Low | Margin of Error | Low |
| 126195 | GB6 Compute Score | 99280 |
| 161.6 img/sec | Background Blur | 145.2 img/sec |
| 117.2 img/sec | Face Detection | 82 img/sec |
| 5.89 Gpixels/sec | Horizon Detection | 3.35 Gpixels/sec |
| 8.05 Gpixels/sec | Edge Detection | 5.01 Gpixels/sec |
| 6.65 Gpixels/sec | Gaussian Blur | 7.59 Gpixels/sec |
| 1.25 Gpixels/sec | Feature Matching | 0.81 Gpixels/sec |
| 550.6 Gpixels/sec | Stereo Matching | 462.4 Gpixels/sec |
| 17613.1 FPS | Particle Physics | 18623.1 FPS |
| OpenCL | API | OpenCL |
| 22128 | G3D Mark Score | 12620 |
| 1005 | G2D Mark | 737 |
| 182 FPS | DirectX 11 | 73 FPS |
| 84 FPS | DirectX 12 | 70 FPS |
| 11062 Ops/s | GPU Compute | 5307 Ops/s |
| 19789 | GB6 ML Single Precision | 12178 |
| 33013 | GB6 ML Half Precision | 12796 |
| 12443 | GB6 ML Quantized | 9086 |
| 7624 | Image Classification (SP) | 5475 |
| 23420 | Image Segmentation (HP) | 13781 |
| 17898 | Image Super Resolution (Q) | 11612 |
| 40535 | Face Detection (HP) | 18556 |
| 83624 | Pose Estimation (Q) | 58801 |
| 2724 | Text Classification (SP) | 824 |
| 4481 | Machine Translation (HP) | 2397 |
| 10304 | Object Detection (SP) | 6895 |
| 30418 | Depth Estimation (Q) | 22219 |
| 180089 | Style Transfer (SP) | 161148 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 166 GPixel/s | Pixel Fill Rate | 269 GPixel/s |
| 317 GTexel/s | Texture Fill Rate | 538 GTexel/s |
| 20.3 TFLOPS | FLOPS (FP32) | 17.2 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 8 GB | Memory Size | 8 GB |
| 1750 MHz | Memory Clock | 2000 MHz |
| 14000 Mbps | Effective Memory Speed | 16000 Mbps |
| 256-bit | Bus | 256-bit |
| No | ECC | No |
| 448 GB/s | Memory Bandwidth | 512 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 220 W | TGP | 225 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 6 nm |
| 392 mm² | Die Size | 406 mm² |
| 17 billion | Transistor Count | 21 billion |
| 43.37 MTr/mm² | Transistor Density | 51.72 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.6 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | No |
| 1.4a | DisplayPort | 2 |
| Nvidia | Vendor | Intel |
| Discrete | Build | Discrete |
| October 22, 2020 | Released | October 12, 2023 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Ampere | Architecture | Generation 12.7 |
| GA104 | GPU Codename | DG2-512 |
| -Intel Core i7 12700Kor above | Recommended CPU | -Intel Core i5 12500or above |
| 14153 | Steel Nomad Lite Score | 10846 |
| 13643 | Time Spy | 12425 |
| 64426 | Solar Bay | 45771 |
| 8311 | Port Royal | 6594 |
| 34054 | Fire Strike | 30444 |
| 28486 | Wild Life Extreme | 23682 |
| 122039 | Night Raid | 110785 |
| 1500 MHz | Base Clock | 2050 MHz |
| 1725 MHz | Boost Clock | 2400 MHz |
| 5888 | Shading Units | 3584 |
| 184 | Texture Mapping Units (TMUs) | 224 |
| 96 | Render Output Units (ROPs) | 112 |
| 46 | Compute Units (Pipelines) | — |
| 184 | Tensor Cores | 448 |
| 46 | Ray-tracing Cores | 28 |
| 128KB per cluster | L1 Cache | — |
| 4MB shared | L2 Cache | 16MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |