Quick/Comparisons
32 | Gaming | 39 |
39 | Workstation | 45 |
33 | AI/ML |
Compare with something else?
Compare GeForce RTX 2070 with another →35 |
27 | Energy Efficiency | 39 |
34 | Quick Comparison Final Score | 40 |
| TU104 | Graphics Processor | GA104 (GN20-E7) |
| 2560 | Cores | 6144 |
| 160 / 64 | TMUs / ROPs | 192 / 96 |
| 2361.66 | Blender GPU | 3258.81 |
| 77 | Forza Horizon 5 | 84 |
| 109 | The Witcher 3 | 127 |
| 150 | Counter-Strike 2 | 165 |
| 91 | Far Cry 6 | 104 |
| 61 | Hogwarts Legacy | 70 |
| 82 | Call of Duty | 98 |
| 58 | Ghost of Tsushima | 66 |
| 71 | Cyberpunk 2077 | 84 |
| 113 | Tomb Raider | 129 |
| 90 | Average FPS | 103 |
| 111 | 1080p High | 125 |
| 90 | 1080p Ultra | 103 |
| 66 | 1440p Ultra | 76 |
| 38 | 4K Ultra | 44 |
| Low | Margin of Error | Low |
| 105349 | GB6 Compute Score | 128827 |
| 166.7 img/sec | Background Blur | 211 img/sec |
| 103.3 img/sec | Face Detection | 123.6 img/sec |
| 4.21 Gpixels/sec | Horizon Detection | 5.78 Gpixels/sec |
| 5.56 Gpixels/sec | Edge Detection | 8.03 Gpixels/sec |
| 5.58 Gpixels/sec | Gaussian Blur | 6.25 Gpixels/sec |
| 1.17 Gpixels/sec | Feature Matching | 1.17 Gpixels/sec |
| 489.7 Gpixels/sec | Stereo Matching | 543.2 Gpixels/sec |
| 13242.9 FPS | Particle Physics | 17774.9 FPS |
| OpenCL | API | OpenCL |
| 18141 | G3D Mark Score | 16147 |
| 878 | G2D Mark | 630 |
| 153 FPS | DirectX 11 | 145 FPS |
| 67 FPS | DirectX 12 | 71 FPS |
| 7473 Ops/s | GPU Compute | 7126 Ops/s |
| 16276 | GB6 ML Single Precision | 18354 |
| 27406 | GB6 ML Half Precision | 29767 |
| 10983 | GB6 ML Quantized | 11487 |
| 6648 | Image Classification (SP) | 6773 |
| 16442 | Image Segmentation (HP) | 19587 |
| 15383 | Image Super Resolution (Q) | 15553 |
| 32861 | Face Detection (HP) | 35885 |
| 66348 | Pose Estimation (Q) | 82446 |
| 2166 | Text Classification (SP) | 2830 |
| 4239 | Machine Translation (HP) | 4019 |
| 8194 | Object Detection (SP) | 8380 |
| 26192 | Depth Estimation (Q) | 28597 |
| 137642 | Style Transfer (SP) | 184715 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 113 GPixel/s | Pixel Fill Rate | 120-164 GPixel/s |
| 283 GTexel/s | Texture Fill Rate | 239-328 GTexel/s |
| 9.1 TFLOPS | FLOPS (FP32) | 15.3-21 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 8 GB | Memory Size | 8-16 GB |
| 1750 MHz | Memory Clock | 1500-1750 MHz |
| 14000 Mbps | Effective Memory Speed | 12000-14000 Mbps |
| 256-bit | Bus | 256-bit |
| No | ECC | No |
| 448 GB/s | Memory Bandwidth | 448 GB/s |
| PCIe 3.0 x16 | Interface | PCIe 4.0 x16 |
| 215 W | TGP | 80-165 W (configurable) |
| TSMC | Manufacturing | Samsung |
| 12 nm | Fabrication Process | 8 nm |
| 545 mm² | Die Size | 392 mm² |
| 13 billion | Transistor Count | 17.4 billion |
| 23.85 MTr/mm² | Transistor Density | 44.39 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.2 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2 |
| 7.5 | CUDA | 8.6 |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | DLSS 2 |
| 1.4a | DisplayPort | 1.4a |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| July 9, 2019 | Released | January 12, 2021 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Turing | Architecture | Ampere |
| TU104 | GPU Codename | GA104 (GN20-E7) |
| -Intel Core i7 9700Kor above | Recommended CPU | -Intel Core i9 12900Hor above |
| 10053 | Steel Nomad Lite Score | 11592 |
| 10167 | Time Spy | 11991 |
| 45871 | Solar Bay | 50371 |
| 6032 | Port Royal | 7201 |
| 25675 | Fire Strike | 29067 |
| 20745 | Wild Life Extreme | 24760 |
| 102728 | Night Raid | 81949 |
| 1605 MHz | Base Clock | 780-1350 MHz (configurable) |
| 1770 MHz | Boost Clock | 1245-1710 MHz (configurable) |
| 2560 | Shading Units | 6144 |
| 160 | Texture Mapping Units (TMUs) | 192 |
| 64 | Render Output Units (ROPs) | 96 |
| 40 | Compute Units (Pipelines) | 48 |
| 320 | Tensor Cores | 192 |
| 40 | Ray-tracing Cores | 48 |
| 64KB per cluster | L1 Cache | 128KB per cluster |
| 4MB shared | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |