Quick/Comparisons
43 | Gaming | 30 |
44 | Workstation | 34 |
35 | AI/ML |
Compare with something else?
Compare GeForce RTX 3070 with another →31 |
35 | Energy Efficiency | 39 |
42 | Quick Comparison Final Score | 31 |
| GA104 | Graphics Processor | Navi 23 |
| 5888 | Cores | 2048 |
| 184 / 96 | TMUs / ROPs | 128 / 64 |
| 3368.42 | Blender GPU | 1138.24 |
| 99 | Forza Horizon 5 | 76 |
| 143 | The Witcher 3 | 101 |
| 194 | Counter-Strike 2 | 130 |
| 121 | Far Cry 6 | 93 |
| 80 | Hogwarts Legacy | 59 |
| 113 | Call of Duty | 87 |
| 74 | Ghost of Tsushima | 63 |
| 94 | Cyberpunk 2077 | 65 |
| 153 | Tomb Raider | 111 |
| 119 | Average FPS | 87 |
| 141 | 1080p High | 112 |
| 119 | 1080p Ultra | 87 |
| 88 | 1440p Ultra | 61 |
| 50 | 4K Ultra | 32 |
| Low | Margin of Error | Low |
| 126195 | GB6 Compute Score | 90831 |
| 161.6 img/sec | Background Blur | 220.1 img/sec |
| 117.2 img/sec | Face Detection | 128.3 img/sec |
| 5.89 Gpixels/sec | Horizon Detection | 2.94 Gpixels/sec |
| 8.05 Gpixels/sec | Edge Detection | 4.13 Gpixels/sec |
| 6.65 Gpixels/sec | Gaussian Blur | 3.68 Gpixels/sec |
| 1.25 Gpixels/sec | Feature Matching | 0.94 Gpixels/sec |
| 550.6 Gpixels/sec | Stereo Matching | 408.2 Gpixels/sec |
| 17613.1 FPS | Particle Physics | 10694.7 FPS |
| OpenCL | API | OpenCL |
| 22128 | G3D Mark Score | 17117 |
| 1005 | G2D Mark | 924 |
| 182 FPS | DirectX 11 | 168 FPS |
| 84 FPS | DirectX 12 | 62 FPS |
| 11062 Ops/s | GPU Compute | 7461 Ops/s |
| 19789 | GB6 ML Single Precision | 17945 |
| 33013 | GB6 ML Half Precision | 21242 |
| 12443 | GB6 ML Quantized | 13940 |
| 7624 | Image Classification (SP) | 8220 |
| 23420 | Image Segmentation (HP) | 9263 |
| 17898 | Image Super Resolution (Q) | 20723 |
| 40535 | Face Detection (HP) | 27074 |
| 83624 | Pose Estimation (Q) | 76541 |
| 2724 | Text Classification (SP) | 2796 |
| 4481 | Machine Translation (HP) | 4281 |
| 10304 | Object Detection (SP) | 6880 |
| 30418 | Depth Estimation (Q) | 33354 |
| 180089 | Style Transfer (SP) | 185358 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 166 GPixel/s | Pixel Fill Rate | 169 GPixel/s |
| 317 GTexel/s | Texture Fill Rate | 337 GTexel/s |
| 20.3 TFLOPS | FLOPS (FP32) | 10.8 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 8 GB | Memory Size | 8 GB |
| 1750 MHz | Memory Clock | 2190 MHz |
| 14000 Mbps | Effective Memory Speed | 17500 Mbps |
| 256-bit | Bus | 128-bit |
| No | ECC | No |
| 448 GB/s | Memory Bandwidth | 280.3 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x8 |
| 220 W | TGP | 176 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 7 nm |
| 392 mm² | Die Size | 237 mm² |
| 17 billion | Transistor Count | 11 billion |
| 43.37 MTr/mm² | Transistor Density | 46.41 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.1 |
| 8.6 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | No |
| 1.4a | DisplayPort | 1.4a |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| October 22, 2020 | Released | May 10, 2022 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Ampere | Architecture | RDNA 2.0 |
| GA104 | GPU Codename | Navi 23 |
| -Intel Core i7 12700Kor above | Recommended CPU | -Intel Core i5 12600Kor above |
| 14153 | Steel Nomad Lite Score | 9265 |
| 13643 | Time Spy | 9897 |
| 64426 | Solar Bay | 38892 |
| 8311 | Port Royal | 4897 |
| 34054 | Fire Strike | 29357 |
| 28486 | Wild Life Extreme | 18034 |
| 122039 | Night Raid | 116256 |
| 1500 MHz | Base Clock | 2055 MHz |
| 1725 MHz | Boost Clock | 2635 MHz |
| 5888 | Shading Units | 2048 |
| 184 | Texture Mapping Units (TMUs) | 128 |
| 96 | Render Output Units (ROPs) | 64 |
| 46 | Compute Units (Pipelines) | — |
| 184 | Tensor Cores | No |
| 46 | Ray-tracing Cores | 32 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 4MB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |