Quick/Comparisons
The GeForce RTX 3050 outperforms the Radeon RX 5500M in gaming due to its higher score of 23, thanks to its Ampere architecture and 8GB GDDR6 memory.
The GeForce RTX 3050 takes the lead in content creation with a score of 29, leveraging its GA106 codename and dedicated AI tensor cores for improved performance.
The GeForce RTX 3050 excels in AI and machine learning tasks with a score of 22, thanks to its dedicated AI tensor cores and support for hardware ray tracing.
23 | Gaming | 13 |
29 | Workstation | 20 |
22 | AI/ML |
Compare with something else?
Compare GeForce RTX 3050 with another →12 |
50 | Energy Efficiency | 64 |
25 | Quick Comparison Final Score | 23 |
| GA106 | Graphics Processor | Navi 14 |
| 2560 | Cores | 1408 |
| 80 / 32 | TMUs / ROPs | 88 / 32 |
| 1475.37 | Blender GPU | 368.1 |
| 52 | Forza Horizon 5 | 39 |
| 71 | The Witcher 3 | 37 |
| 99 | Counter-Strike 2 | 48 |
| 68 | Far Cry 6 | 46 |
| 37 | Hogwarts Legacy | 25 |
| 52 | Call of Duty | 40 |
| 35 | Ghost of Tsushima | 20 |
| 42 | Cyberpunk 2077 | 27 |
| 76 | Tomb Raider | 46 |
| 59 | Average FPS | 36 |
| 75 | 1080p High | 48 |
| 59 | 1080p Ultra | 36 |
| 43 | 1440p Ultra | 24 |
| 23 | 4K Ultra | 13 |
| Low | Margin of Error | Low |
| 67028 | GB6 Compute Score | 46415 |
| 110.4 img/sec | Background Blur | 89.3 img/sec |
| 64.4 img/sec | Face Detection | 50.3 img/sec |
| 2.87 Gpixels/sec | Horizon Detection | 1.91 Gpixels/sec |
| 3.76 Gpixels/sec | Edge Detection | 2.74 Gpixels/sec |
| 2.89 Gpixels/sec | Gaussian Blur | 2.25 Gpixels/sec |
| 0.81 Gpixels/sec | Feature Matching | 0.49 Gpixels/sec |
| 270.9 Gpixels/sec | Stereo Matching | 159.9 Gpixels/sec |
| 9339.7 FPS | Particle Physics | 5898.4 FPS |
| OpenCL | API | OpenCL |
| 12503 | G3D Mark Score | 5998 |
| 918 | G2D Mark | 413 |
| 90 FPS | DirectX 11 | 36 FPS |
| 59 FPS | DirectX 12 | 29 FPS |
| 6438 Ops/s | GPU Compute | 2338 Ops/s |
| 11848 | GB6 ML Single Precision | 6824 |
| 21368 | GB6 ML Half Precision | 7661 |
| 7905 | GB6 ML Quantized | 5150 |
| 4958 | Image Classification (SP) | 2845 |
| 12747 | Image Segmentation (HP) | 3023 |
| 9816 | Image Super Resolution (Q) | 8505 |
| 22467 | Face Detection (HP) | 8964 |
| 41362 | Pose Estimation (Q) | 29278 |
| 2563 | Text Classification (SP) | 1142 |
| 3721 | Machine Translation (HP) | 1920 |
| 5929 | Object Detection (SP) | 3032 |
| 18361 | Depth Estimation (Q) | 14009 |
| 84665 | Style Transfer (SP) | 58664 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 57 GPixel/s | Pixel Fill Rate | 53 GPixel/s |
| 142 GTexel/s | Texture Fill Rate | 145 GTexel/s |
| 9.1 TFLOPS | FLOPS (FP32) | 4.6 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 8 GB | Memory Size | 4 GB |
| 1750 MHz | Memory Clock | 1750 MHz |
| 14000 Mbps | Effective Memory Speed | 14000 Mbps |
| 128-bit | Bus | 128-bit |
| No | ECC | — |
| 224 GB/s | Memory Bandwidth | 224 GB/s |
| PCIe 4.0 x8 | Interface | PCIe 4.0 x8 |
| 130 W | TGP | 85 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 7 nm |
| 276 mm² | Die Size | 158 mm² |
| 12 billion | Transistor Count | 6.4 billion |
| 43.48 MTr/mm² | Transistor Density | 40.51 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.2 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.1 |
| 8.6 | CUDA | — |
| Yes | Ray Tracing | No |
| DLSS 2 | DLSS | No |
| 1.4a | DisplayPort | 1.4 |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| January 27, 2022 | Released | October 7, 2019 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Gaming |
| Entry-level | Segment | High-end |
| Ampere | Architecture | RDNA 1.0 |
| GA106 | GPU Codename | Navi 14 |
| -Intel Core i5 12400or above | Recommended CPU | -AMD Ryzen 5 5600Hor above |
| 6352 | Steel Nomad Lite Score | 3774 |
| 6235 | Time Spy | 4292 |
| 30139 | Solar Bay | — |
| 3545 | Port Royal | — |
| 15968 | Fire Strike | 12807 |
| 12803 | Wild Life Extreme | 6440 |
| 74799 | Night Raid | 42033 |
| 1552 MHz | Base Clock | 1375 MHz |
| 1777 MHz | Boost Clock | 1645 MHz |
| 2560 | Shading Units | 1408 |
| 80 | Texture Mapping Units (TMUs) | 88 |
| 32 | Render Output Units (ROPs) | 32 |
| 20 | Compute Units (Pipelines) | 22 |
| 80 | Tensor Cores | No |
| 20 | Ray-tracing Cores | No |
| 128KB per cluster | L1 Cache | — |
| 2MB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |