Quick/Comparisons
43 | Gaming | 34 |
44 | Workstation | 41 |
35 | AI/ML |
Compare with something else?
Compare GeForce RTX 3070 with another →38 |
35 | Energy Efficiency | 47 |
42 | Quick Comparison Final Score | 37 |
| GA104 | Graphics Processor | BMG-G21 |
| 5888 | Cores | 2560 |
| 184 / 96 | TMUs / ROPs | 160 / 80 |
| 3368.42 | Blender GPU | 2106.04 |
| 99 | Forza Horizon 5 | 78 |
| 143 | The Witcher 3 | 128 |
| 194 | Counter-Strike 2 | 116 |
| 121 | Far Cry 6 | 98 |
| 80 | Hogwarts Legacy | 57 |
| 113 | Call of Duty | 85 |
| 74 | Ghost of Tsushima | 61 |
| 94 | Cyberpunk 2077 | 82 |
| 153 | Tomb Raider | 107 |
| 119 | Average FPS | 90 |
| 141 | 1080p High | 109 |
| 119 | 1080p Ultra | 90 |
| 88 | 1440p Ultra | 67 |
| 50 | 4K Ultra | 38 |
| Low | Margin of Error | Low |
| 126195 | GB6 Compute Score | 111561 |
| 161.6 img/sec | Background Blur | 284.6 img/sec |
| 117.2 img/sec | Face Detection | 150 img/sec |
| 5.89 Gpixels/sec | Horizon Detection | 2.67 Gpixels/sec |
| 8.05 Gpixels/sec | Edge Detection | 5.22 Gpixels/sec |
| 6.65 Gpixels/sec | Gaussian Blur | 6.35 Gpixels/sec |
| 1.25 Gpixels/sec | Feature Matching | 1.18 Gpixels/sec |
| 550.6 Gpixels/sec | Stereo Matching | 397 Gpixels/sec |
| 17613.1 FPS | Particle Physics | 15177.3 FPS |
| OpenCL | API | OpenCL |
| 22128 | G3D Mark Score | 15914 |
| 1005 | G2D Mark | 700 |
| 182 FPS | DirectX 11 | 126 FPS |
| 84 FPS | DirectX 12 | 77 FPS |
| 11062 Ops/s | GPU Compute | 7896 Ops/s |
| 19789 | GB6 ML Single Precision | 21661 |
| 33013 | GB6 ML Half Precision | 34463 |
| 12443 | GB6 ML Quantized | 14366 |
| 7624 | Image Classification (SP) | 10654 |
| 23420 | Image Segmentation (HP) | 18608 |
| 17898 | Image Super Resolution (Q) | 16715 |
| 40535 | Face Detection (HP) | 40348 |
| 83624 | Pose Estimation (Q) | 64494 |
| 2724 | Text Classification (SP) | 3091 |
| 4481 | Machine Translation (HP) | 3530 |
| 10304 | Object Detection (SP) | 13690 |
| 30418 | Depth Estimation (Q) | 36797 |
| 180089 | Style Transfer (SP) | 197057 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 166 GPixel/s | Pixel Fill Rate | 214 GPixel/s |
| 317 GTexel/s | Texture Fill Rate | 427 GTexel/s |
| 20.3 TFLOPS | FLOPS (FP32) | 13.7 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 8 GB | Memory Size | 12 GB |
| 1750 MHz | Memory Clock | 2375 MHz |
| 14000 Mbps | Effective Memory Speed | 19000 Mbps |
| 256-bit | Bus | 192-bit |
| No | ECC | No |
| 448 GB/s | Memory Bandwidth | 456 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x8 |
| 220 W | TGP | 190 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 5 nm |
| 392 mm² | Die Size | 272 mm² |
| 17 billion | Transistor Count | 19.6 billion |
| 43.37 MTr/mm² | Transistor Density | 72.06 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.6 | CUDA | No |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | No |
| 1.4a | DisplayPort | 2.1 |
| Nvidia | Vendor | Intel |
| Discrete | Build | Discrete |
| October 22, 2020 | Released | December 3, 2024 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Ampere | Architecture | Generation 12.7 |
| GA104 | GPU Codename | BMG-G21 |
| -Intel Core i7 12700Kor above | Recommended CPU | -Intel Core Ultra 5 245Kor above |
| 14153 | Steel Nomad Lite Score | 11180 |
| 13643 | Time Spy | 14665 |
| 64426 | Solar Bay | 55557 |
| 8311 | Port Royal | 7919 |
| 34054 | Fire Strike | 36685 |
| 28486 | Wild Life Extreme | 28324 |
| 122039 | Night Raid | 123864 |
| 1500 MHz | Base Clock | 1700 MHz |
| 1725 MHz | Boost Clock | 2670 MHz |
| 5888 | Shading Units | 2560 |
| 184 | Texture Mapping Units (TMUs) | 160 |
| 96 | Render Output Units (ROPs) | 80 |
| 46 | Compute Units (Pipelines) | 20 |
| 184 | Tensor Cores | 160 |
| 46 | Ray-tracing Cores | 20 |
| 128KB per cluster | L1 Cache | 256KB per cluster |
| 4MB shared | L2 Cache | 12MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |