Quick/Comparisons
37 | Gaming | 47 |
41 | Workstation | 52 |
31 | AI/ML |
Compare with something else?
Compare GeForce RTX 3070 with another →44 |
47 | Energy Efficiency | 59 |
39 | Quick Comparison Final Score | 48 |
| GA104 (GN20-E6) | Graphics Processor | AD104 |
| 5632 | Cores | 6144 |
| 176 / 80 | TMUs / ROPs | 192 / 80 |
| 3120.06 | Blender GPU | 4481.44 |
| 85 | Forza Horizon 5 | 98 |
| 123 | The Witcher 3 | 143 |
| 167 | Counter-Strike 2 | 201 |
| 101 | Far Cry 6 | 123 |
| 67 | Hogwarts Legacy | 82 |
| 96 | Call of Duty | 119 |
| 63 | Ghost of Tsushima | 79 |
| 81 | Cyberpunk 2077 | 96 |
| 126 | Tomb Raider | 157 |
| 101 | Average FPS | 122 |
| 122 | 1080p High | 147 |
| 101 | 1080p Ultra | 122 |
| 74 | 1440p Ultra | 92 |
| 42 | 4K Ultra | 53 |
| Low | Margin of Error | High |
| 116337 | GB6 Compute Score | 151680 |
| 173.8 img/sec | Background Blur | 235.3 img/sec |
| 96.9 img/sec | Face Detection | 157.9 img/sec |
| 5.69 Gpixels/sec | Horizon Detection | 5.53 Gpixels/sec |
| 7.75 Gpixels/sec | Edge Detection | 7.44 Gpixels/sec |
| 6.27 Gpixels/sec | Gaussian Blur | 7.87 Gpixels/sec |
| 1.14 Gpixels/sec | Feature Matching | 1.58 Gpixels/sec |
| 480.4 Gpixels/sec | Stereo Matching | 737.7 Gpixels/sec |
| 14706.2 FPS | Particle Physics | 22409.2 FPS |
| OpenCL | API | OpenCL |
| 17472 | G3D Mark Score | 23716 |
| 773 | G2D Mark | 1118 |
| 152 FPS | DirectX 11 | 204 FPS |
| 72 FPS | DirectX 12 | 93 FPS |
| 7362 Ops/s | GPU Compute | 11936 Ops/s |
| 17250 | GB6 ML Single Precision | 22372 |
| 27936 | GB6 ML Half Precision | 35651 |
| 11249 | GB6 ML Quantized | 17081 |
| 6969 | Image Classification (SP) | 9717 |
| 17142 | Image Segmentation (HP) | 19365 |
| 16426 | Image Super Resolution (Q) | 29227 |
| 31628 | Face Detection (HP) | 40416 |
| 74413 | Pose Estimation (Q) | 105639 |
| 2522 | Text Classification (SP) | 3577 |
| 3639 | Machine Translation (HP) | 4193 |
| 9002 | Object Detection (SP) | 10820 |
| 27790 | Depth Estimation (Q) | 36574 |
| 160005 | Style Transfer (SP) | 200936 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 83-119 GPixel/s | Pixel Fill Rate | 174 GPixel/s |
| 182-261 GTexel/s | Texture Fill Rate | 418 GTexel/s |
| 11.7-16.7 TFLOPS | FLOPS (FP32) | 26.7 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 8 GB | Memory Size | 20 GB |
| 1750 MHz | Memory Clock | 2250 MHz |
| 14000 Mbps | Effective Memory Speed | 18000 Mbps |
| 256-bit | Bus | 160-bit |
| No | ECC | No |
| 448 GB/s | Memory Bandwidth | 360 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 80-150 W (configurable) | TGP | 130 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 5 nm |
| 392 mm² | Die Size | 294 mm² |
| 17.4 billion | Transistor Count | 35 billion |
| 44.39 MTr/mm² | Transistor Density | 119.05 MTr/mm² |
| 12 | DirectX | 12 |
| 1.2 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.6 | CUDA | 8.9 |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | DLSS 3 |
| 1.4a | DisplayPort | 1.4a |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| — | Released | August 9, 2023 |
| Laptop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | High-end |
| Ampere | Architecture | Ada Lovelace |
| GA104 (GN20-E6) | GPU Codename | AD104 |
| -Intel Core i7 12700Hor above | Recommended CPU | -Intel Core i7 14700Kor above |
| 11488 | Steel Nomad Lite Score | 14674 |
| 11331 | Time Spy | 14438 |
| 49158 | Solar Bay | 65875 |
| 6850 | Port Royal | 8955 |
| 28619 | Fire Strike | 37363 |
| 23071 | Wild Life Extreme | — |
| 95266 | Night Raid | 158480 |
| 510-1035 MHz (configurable) | Base Clock | 1500 MHz |
| 1035-1485 MHz (configurable) | Boost Clock | 2175 MHz |
| 5632 | Shading Units | 6144 |
| 176 | Texture Mapping Units (TMUs) | 192 |
| 80 | Render Output Units (ROPs) | 80 |
| 44 | Compute Units (Pipelines) | 48 |
| 176 | Tensor Cores | 192 |
| 44 | Ray-tracing Cores | 48 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 4MB shared | L2 Cache | 48MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |