Quick/Comparisons
25 | Gaming | 13 |
35 | Workstation | 22 |
24 | AI/ML |
Compare with something else?
Compare GeForce RTX 2070 with another →12 |
46 | Energy Efficiency | 32 |
31 | Quick Comparison Final Score | 16 |
| TU106B | Graphics Processor | Polaris 20 |
| 2304 | Cores | 2048 |
| 144 / 64 | TMUs / ROPs | 128 / 32 |
| 1823.68 | Blender GPU | 329.21 |
| 61 | Forza Horizon 5 | 36 |
| 84 | The Witcher 3 | 34 |
| 117 | Counter-Strike 2 | 43 |
| 74 | Far Cry 6 | 40 |
| 45 | Hogwarts Legacy | 22 |
| 61 | Call of Duty | 36 |
| 42 | Ghost of Tsushima | 21 |
| 53 | Cyberpunk 2077 | 26 |
| 84 | Tomb Raider | 42 |
| 69 | Average FPS | 33 |
| 87 | 1080p High | 45 |
| 69 | 1080p Ultra | 33 |
| 49 | 1440p Ultra | 23 |
| 26 | 4K Ultra | 12 |
| Low | Margin of Error | Low |
| 92843 | GB6 Compute Score | 37147 |
| 159 img/sec | Background Blur | 55.7 img/sec |
| 91.2 img/sec | Face Detection | 31.3 img/sec |
| 3.67 Gpixels/sec | Horizon Detection | 1.41 Gpixels/sec |
| 5.1 Gpixels/sec | Edge Detection | 2.82 Gpixels/sec |
| 5.03 Gpixels/sec | Gaussian Blur | 2.57 Gpixels/sec |
| 1.08 Gpixels/sec | Feature Matching | 0.31 Gpixels/sec |
| 411.9 Gpixels/sec | Stereo Matching | 143.2 Gpixels/sec |
| 10180.5 FPS | Particle Physics | 5080.1 FPS |
| OpenCL | API | OpenCL |
| 12357 | G3D Mark Score | 7922 |
| 602 | G2D Mark | 724 |
| 111 FPS | DirectX 11 | 55 FPS |
| 41 FPS | DirectX 12 | 39 FPS |
| 4490 Ops/s | GPU Compute | 2890 Ops/s |
| 10388 | GB6 ML Single Precision | 6885 |
| 17021 | GB6 ML Half Precision | 7446 |
| 7357 | GB6 ML Quantized | 4931 |
| 3954 | Image Classification (SP) | 2676 |
| 9425 | Image Segmentation (HP) | 3757 |
| 11209 | Image Super Resolution (Q) | 6179 |
| 18147 | Face Detection (HP) | 10752 |
| 47333 | Pose Estimation (Q) | 35488 |
| 1546 | Text Classification (SP) | 1051 |
| 2209 | Machine Translation (HP) | 1933 |
| 4976 | Object Detection (SP) | 3048 |
| 18466 | Depth Estimation (Q) | 12040 |
| 90152 | Style Transfer (SP) | 71201 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 92-95 GPixel/s | Pixel Fill Rate | 40 GPixel/s |
| 207-214 GTexel/s | Texture Fill Rate | 159 GTexel/s |
| 6.6-6.8 TFLOPS | FLOPS (FP32) | 5.1 TFLOPS |
| GDDR6 | Memory Type | GDDR5 |
| 8 GB | Memory Size | 4 GB |
| 1750 MHz | Memory Clock | 1750 MHz |
| 14000 Mbps | Effective Memory Speed | 7000 Mbps |
| 256-bit | Bus | 256-bit |
| No | ECC | No |
| 448 GB/s | Memory Bandwidth | 224 GB/s |
| PCIe 3.0 x16 | Interface | PCIe 3.0 x16 |
| 115 W | TGP | 150 W |
| TSMC | Manufacturing | GlobalFoundries |
| 12 nm | Fabrication Process | 14 nm |
| 445 mm² | Die Size | 232 mm² |
| 10.8 billion | Transistor Count | 5 billion |
| 24.27 MTr/mm² | Transistor Density | 21.55 MTr/mm² |
| 12 | DirectX | 12 |
| 1.2 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 1.2 | OpenCL | 2.1 |
| 7.5 | CUDA | — |
| Yes | Ray Tracing | No |
| DLSS 2 | DLSS | No |
| 1.4 | DisplayPort | 1.4a |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| January 29, 2019 | Released | April 18, 2017 |
| Laptop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | Mid-range |
| Turing | Architecture | GCN 4.0 |
| TU106B | GPU Codename | Polaris 20 |
| -Intel Core i7 10750Hor above | Recommended CPU | -Intel Core i5 8500or above |
| 7234 | Steel Nomad Lite Score | 3702 |
| 7566 | Time Spy | 3855 |
| 31700 | Solar Bay | — |
| 4446 | Port Royal | — |
| 19887 | Fire Strike | 13443 |
| 14527 | Wild Life Extreme | 7941 |
| 65976 | Night Raid | 50379 |
| 1215-1305 MHz (configurable) | Base Clock | 1168 MHz |
| 1440-1485 MHz (configurable) | Boost Clock | 1244 MHz |
| 2304 | Shading Units | 2048 |
| 144 | Texture Mapping Units (TMUs) | 128 |
| 64 | Render Output Units (ROPs) | 32 |
| 36 | Compute Units (Pipelines) | — |
| 288 | Tensor Cores | No |
| 36 | Ray-tracing Cores | No |
| 64KB per cluster | L1 Cache | 16KB per cluster |
| 4MB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |