Quick/Comparisons
25 | Gaming | 31 |
35 | Workstation | 37 |
24 | AI/ML |
Compare with something else?
Compare GeForce RTX 2070 with another →29 |
46 | Energy Efficiency | 37 |
31 | Quick Comparison Final Score | 33 |
| TU106B | Graphics Processor | GA106 |
| 2304 | Cores | 3584 |
| 144 / 64 | TMUs / ROPs | 112 / 48 |
| 1823.68 | Blender GPU | 2303.56 |
| 61 | Forza Horizon 5 | 71 |
| 84 | The Witcher 3 | 100 |
| 117 | Counter-Strike 2 | 133 |
| 74 | Far Cry 6 | 83 |
| 45 | Hogwarts Legacy | 56 |
| 61 | Call of Duty | 73 |
| 42 | Ghost of Tsushima | 49 |
| 53 | Cyberpunk 2077 | 62 |
| 84 | Tomb Raider | 97 |
| 69 | Average FPS | 80 |
| 87 | 1080p High | 100 |
| 69 | 1080p Ultra | 80 |
| 49 | 1440p Ultra | 59 |
| 26 | 4K Ultra | 32 |
| Low | Margin of Error | Low |
| 92843 | GB6 Compute Score | 91603 |
| 159 img/sec | Background Blur | 149.6 img/sec |
| 91.2 img/sec | Face Detection | 91 img/sec |
| 3.67 Gpixels/sec | Horizon Detection | 4.07 Gpixels/sec |
| 5.1 Gpixels/sec | Edge Detection | 5.58 Gpixels/sec |
| 5.03 Gpixels/sec | Gaussian Blur | 3.92 Gpixels/sec |
| 1.08 Gpixels/sec | Feature Matching | 1.01 Gpixels/sec |
| 411.9 Gpixels/sec | Stereo Matching | 372.9 Gpixels/sec |
| 10180.5 FPS | Particle Physics | 12226.3 FPS |
| OpenCL | API | OpenCL |
| 12357 | G3D Mark Score | 16744 |
| 602 | G2D Mark | 966 |
| 111 FPS | DirectX 11 | 128 FPS |
| 41 FPS | DirectX 12 | 69 FPS |
| 4490 Ops/s | GPU Compute | 8266 Ops/s |
| 10388 | GB6 ML Single Precision | 15123 |
| 17021 | GB6 ML Half Precision | 25942 |
| 7357 | GB6 ML Quantized | 10356 |
| 3954 | Image Classification (SP) | 6645 |
| 9425 | Image Segmentation (HP) | 15338 |
| 11209 | Image Super Resolution (Q) | 14283 |
| 18147 | Face Detection (HP) | 28862 |
| 47333 | Pose Estimation (Q) | 61313 |
| 1546 | Text Classification (SP) | 2463 |
| 2209 | Machine Translation (HP) | 3793 |
| 4976 | Object Detection (SP) | 7712 |
| 18466 | Depth Estimation (Q) | 25461 |
| 90152 | Style Transfer (SP) | 123233 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 92-95 GPixel/s | Pixel Fill Rate | 85 GPixel/s |
| 207-214 GTexel/s | Texture Fill Rate | 199 GTexel/s |
| 6.6-6.8 TFLOPS | FLOPS (FP32) | 12.7 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 8 GB | Memory Size | 12 GB |
| 1750 MHz | Memory Clock | 1875 MHz |
| 14000 Mbps | Effective Memory Speed | 15000 Mbps |
| 256-bit | Bus | 192-bit |
| No | ECC | No |
| 448 GB/s | Memory Bandwidth | 360 GB/s |
| PCIe 3.0 x16 | Interface | PCIe 4.0 x16 |
| 115 W | TGP | 170 W |
| TSMC | Manufacturing | Samsung |
| 12 nm | Fabrication Process | 8 nm |
| 445 mm² | Die Size | 276 mm² |
| 10.8 billion | Transistor Count | 12 billion |
| 24.27 MTr/mm² | Transistor Density | 43.48 MTr/mm² |
| 12 | DirectX | 12 |
| 1.2 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 1.2 | OpenCL | 3 |
| 7.5 | CUDA | 8.6 |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | DLSS 2 |
| 1.4 | DisplayPort | 1.4a |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| January 29, 2019 | Released | January 12, 2021 |
| Laptop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | Mid-range |
| Turing | Architecture | Ampere |
| TU106B | GPU Codename | GA106 |
| -Intel Core i7 10750Hor above | Recommended CPU | -Intel Core i5 12600Kor above |
| 7234 | Steel Nomad Lite Score | 9002 |
| 7566 | Time Spy | 8754 |
| 31700 | Solar Bay | 41139 |
| 4446 | Port Royal | 5145 |
| 19887 | Fire Strike | 22306 |
| 14527 | Wild Life Extreme | 17726 |
| 65976 | Night Raid | 97592 |
| 1215-1305 MHz (configurable) | Base Clock | 1320 MHz |
| 1440-1485 MHz (configurable) | Boost Clock | 1777 MHz |
| 2304 | Shading Units | 3584 |
| 144 | Texture Mapping Units (TMUs) | 112 |
| 64 | Render Output Units (ROPs) | 48 |
| 36 | Compute Units (Pipelines) | 28 |
| 288 | Tensor Cores | 112 |
| 36 | Ray-tracing Cores | 28 |
| 64KB per cluster | L1 Cache | 128KB per cluster |
| 4MB shared | L2 Cache | 3MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |