Quick/Comparisons
40 | Gaming | 12 |
40 | Workstation | 23 |
31 | AI/ML |
Compare with something else?
Compare Nvidia RTX 2000 with another →14 |
74 | Energy Efficiency | 72 |
39 | Quick Comparison Final Score | 25 |
| AD107 | Graphics Processor | GA107S |
| 2816 | Cores | 2048 |
| 88 / 48 | TMUs / ROPs | 64 / 32 |
| 2690.32 | Blender GPU | 692.94 |
| — | Forza Horizon 5 | 24 |
| — | The Witcher 3 | 26 |
| — | Counter-Strike 2 | 42 |
| — | Far Cry 6 | 33 |
| — | Hogwarts Legacy | 14 |
| — | Call of Duty | 26 |
| — | Ghost of Tsushima | 13 |
| — | Cyberpunk 2077 | 18 |
| — | Tomb Raider | 36 |
| — | Average FPS | 26 |
| — | 1080p High | 35 |
| — | 1080p Ultra | 26 |
| — | 1440p Ultra | 18 |
| — | 4K Ultra | 9 |
| — | Margin of Error | Medium |
| 89677 | GB6 Compute Score | 40574 |
| 170.5 img/sec | Background Blur | 62.4 img/sec |
| 102.8 img/sec | Face Detection | 35.7 img/sec |
| 3.25 Gpixels/sec | Horizon Detection | 1.62 Gpixels/sec |
| 4.12 Gpixels/sec | Edge Detection | 2.11 Gpixels/sec |
| 4.06 Gpixels/sec | Gaussian Blur | 2.05 Gpixels/sec |
| 1.13 Gpixels/sec | Feature Matching | 0.52 Gpixels/sec |
| 353.4 Gpixels/sec | Stereo Matching | 185.1 Gpixels/sec |
| 12273.1 FPS | Particle Physics | 5502.5 FPS |
| OpenCL | API | OpenCL |
| 17146 | G3D Mark Score | 6570 |
| 1094 | G2D Mark | 416 |
| 140 FPS | DirectX 11 | 40 FPS |
| 72 FPS | DirectX 12 | 38 FPS |
| 7990 Ops/s | GPU Compute | 2911 Ops/s |
| 15501 | GB6 ML Single Precision | 7597 |
| 27554 | GB6 ML Half Precision | 15030 |
| 12117 | GB6 ML Quantized | 5337 |
| 6382 | Image Classification (SP) | 3298 |
| 16489 | Image Segmentation (HP) | 8358 |
| 17694 | Image Super Resolution (Q) | 6219 |
| 32453 | Face Detection (HP) | 14319 |
| 56937 | Pose Estimation (Q) | 22995 |
| 3097 | Text Classification (SP) | 2226 |
| 4324 | Machine Translation (HP) | 3163 |
| 7656 | Object Detection (SP) | 3768 |
| 29700 | Depth Estimation (Q) | 12079 |
| 116441 | Style Transfer (SP) | 53139 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 102 GPixel/s | Pixel Fill Rate | 24-55 GPixel/s |
| 187 GTexel/s | Texture Fill Rate | 48-109 GTexel/s |
| 12 TFLOPS | FLOPS (FP32) | 3.1-7 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 16 GB | Memory Size | 4 GB |
| 2000 MHz | Memory Clock | 1500 MHz |
| 16000 Mbps | Effective Memory Speed | 12000 Mbps |
| 128-bit | Bus | 64-bit |
| No | ECC | No |
| 256 GB/s | Memory Bandwidth | 96 GB/s |
| PCIe 4.0 x8 | Interface | PCIe 4.0 x8 |
| 70 W | TGP | 20-60 W (configurable) |
| TSMC | Manufacturing | Samsung |
| 5 nm | Fabrication Process | 8 nm |
| 159 mm² | Die Size | 200 mm² |
| 18 billion | Transistor Count | 8.7 billion |
| 113.21 MTr/mm² | Transistor Density | 43.5 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.9 | CUDA | 8.6 |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | DLSS 2 |
| 1.4a | DisplayPort | 1.4a |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| February 12, 2022 | Released | March 22, 2022 |
| Desktop | Case | Laptop |
| Professional | Purpose | Professional |
| Mid-range | Segment | Entry-level |
| Ada Lovelace | Architecture | Ampere |
| AD107 | GPU Codename | GA107S |
| -Intel Core i7 14700Kor above | Recommended CPU | -Intel Core i7 12700Hor above |
| — | Steel Nomad Lite Score | 3021 |
| — | Time Spy | 3023 |
| — | Solar Bay | 13710 |
| — | Port Royal | 625 |
| — | Fire Strike | 10265 |
| — | Wild Life Extreme | 4053 |
| — | Night Raid | 38312 |
| 1620 MHz | Base Clock | 370-890 MHz (configurable) |
| 2130 MHz | Boost Clock | 745-1710 MHz (configurable) |
| 2816 | Shading Units | 2048 |
| 88 | Texture Mapping Units (TMUs) | 64 |
| 48 | Render Output Units (ROPs) | 32 |
| 22 | Compute Units (Pipelines) | 16 |
| 88 | Tensor Cores | 64 |
| 22 | Ray-tracing Cores | 16 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 12MB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |