Quick/Comparisons
The GeForce MX570 takes the lead in gaming due to its higher score of 21, thanks to its Ampere architecture and 4GB GDDR6 memory.
The Nvidia RTX A2000 Laptop outperforms the GeForce MX570 in content creation with a score of 29, driven by its strong workstation capabilities.
The Nvidia RTX A2000 Laptop takes the lead in AI & Machine Learning with a score of 18, thanks to its dedicated AI tensor cores and Ampere architecture.
21 | Gaming | 17 |
17 | Workstation | 29 |
12 | AI/ML |
Compare with something else?
Compare GeForce MX570 with another →18 |
75 | Energy Efficiency | 59 |
28 | Quick Comparison Final Score | 27 |
| GA107S | Graphics Processor | GA106 |
| 2048 | Cores | 2560 |
| 64 / 40 | TMUs / ROPs | 80 / 48 |
| 489.72 | Blender GPU | 1262.73 |
| — | Forza Horizon 5 | 44 |
| — | The Witcher 3 | 47 |
| — | Counter-Strike 2 | 74 |
| — | Far Cry 6 | 50 |
| — | Hogwarts Legacy | 26 |
| — | Call of Duty | 36 |
| — | Ghost of Tsushima | 23 |
| — | Cyberpunk 2077 | 33 |
| — | Tomb Raider | 56 |
| — | Average FPS | 43 |
| — | 1080p High | 55 |
| — | 1080p Ultra | 43 |
| — | 1440p Ultra | 30 |
| — | 4K Ultra | 14 |
| — | Margin of Error | Low |
| 42999 | GB6 Compute Score | 57201 |
| 77.2 img/sec | Background Blur | 92.8 img/sec |
| 40.3 img/sec | Face Detection | 54.2 img/sec |
| 1.65 Gpixels/sec | Horizon Detection | 2.47 Gpixels/sec |
| 2.13 Gpixels/sec | Edge Detection | 3.32 Gpixels/sec |
| 2.04 Gpixels/sec | Gaussian Blur | 2.63 Gpixels/sec |
| 0.57 Gpixels/sec | Feature Matching | 0.7 Gpixels/sec |
| 181.5 Gpixels/sec | Stereo Matching | 236.9 Gpixels/sec |
| 5728.6 FPS | Particle Physics | 7148.8 FPS |
| OpenCL | API | OpenCL |
| 5795 | G3D Mark Score | 9487 |
| 388 | G2D Mark | 505 |
| 38 FPS | DirectX 11 | 65 FPS |
| 30 FPS | DirectX 12 | 47 FPS |
| 2561 Ops/s | GPU Compute | 4252 Ops/s |
| 5997 | GB6 ML Single Precision | 9891 |
| 11591 | GB6 ML Half Precision | 18270 |
| 4316 | GB6 ML Quantized | 6987 |
| 2792 | Image Classification (SP) | 4240 |
| 5831 | Image Segmentation (HP) | 9190 |
| 5115 | Image Super Resolution (Q) | 9374 |
| 9491 | Face Detection (HP) | 18335 |
| 19218 | Pose Estimation (Q) | 36989 |
| 1996 | Text Classification (SP) | 2035 |
| 1992 | Machine Translation (HP) | 2687 |
| 2848 | Object Detection (SP) | 5174 |
| 10712 | Depth Estimation (Q) | 16537 |
| 44344 | Style Transfer (SP) | 77854 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 59 GPixel/s | Pixel Fill Rate | 50-81 GPixel/s |
| 95 GTexel/s | Texture Fill Rate | 83-136 GTexel/s |
| 6 TFLOPS | FLOPS (FP32) | 5.3-8.7 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 4 GB | Memory Size | 4 GB |
| 1500 MHz | Memory Clock | 1500 MHz |
| 12000 Mbps | Effective Memory Speed | 12000 Mbps |
| 64-bit | Bus | 128-bit |
| — | ECC | No |
| 96 GB/s | Memory Bandwidth | 192 GB/s |
| PCIe 4.0 x8 | Interface | PCIe 4.0 x16 |
| 45 W | TGP | 60-95 W (configurable) |
| Samsung | Manufacturing | Samsung |
| 8 nm | Fabrication Process | 8 nm |
| 200 mm² | Die Size | 276 mm² |
| 8.7 billion | Transistor Count | 12 billion |
| 43.5 MTr/mm² | Transistor Density | 43.48 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.6 | CUDA | 8.6 |
| Yes | Ray Tracing | Yes |
| No | DLSS | DLSS 2 |
| 1.4 | DisplayPort | 1.4a |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| January 1, 2022 | Released | April 12, 2021 |
| Laptop | Case | Laptop |
| Gaming | Purpose | Professional |
| Entry-level | Segment | Mid-range |
| Ampere | Architecture | Ampere |
| GA107S | GPU Codename | GA106 |
| -Intel Core i7 1255Uor above | Recommended CPU | -Intel Core i7 12800Hor above |
| — | Steel Nomad Lite Score | 4397 |
| — | Time Spy | 4723 |
| — | Solar Bay | 19912 |
| — | Port Royal | 2895 |
| — | Fire Strike | 11357 |
| — | Wild Life Extreme | 8429 |
| — | Night Raid | 49889 |
| 1155 MHz | Base Clock | 735-1463 MHz (configurable) |
| 1477 MHz | Boost Clock | 1035-1695 MHz (configurable) |
| 2048 | Shading Units | 2560 |
| 64 | Texture Mapping Units (TMUs) | 80 |
| 40 | Render Output Units (ROPs) | 48 |
| 16 | Compute Units (Pipelines) | 20 |
| 64 | Tensor Cores | 80 |
| 16 | Ray-tracing Cores | 20 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 2MB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |