Quick/Comparisons
23 | Gaming | 28 |
29 | Workstation | 30 |
22 | AI/ML |
Compare with something else?
Compare GeForce RTX 3050 with another →27 |
50 | Energy Efficiency | 56 |
25 | Quick Comparison Final Score | 33 |
| GA106 | Graphics Processor | AD107 (GN21-X2) |
| 2560 | Cores | 2560 |
| 80 / 32 | TMUs / ROPs | 80 / 32 |
| 1475.37 | Blender GPU | 2575.82 |
| 52 | Forza Horizon 5 | 66 |
| 71 | The Witcher 3 | 92 |
| 99 | Counter-Strike 2 | 125 |
| 68 | Far Cry 6 | 79 |
| 37 | Hogwarts Legacy | 51 |
| 52 | Call of Duty | 64 |
| 35 | Ghost of Tsushima | 43 |
| 42 | Cyberpunk 2077 | 56 |
| 76 | Tomb Raider | 93 |
| 59 | Average FPS | 74 |
| 75 | 1080p High | 93 |
| 59 | 1080p Ultra | 74 |
| 43 | 1440p Ultra | 54 |
| 23 | 4K Ultra | 30 |
| Low | Margin of Error | Low |
| 67028 | GB6 Compute Score | 81436 |
| 110.4 img/sec | Background Blur | 141.4 img/sec |
| 64.4 img/sec | Face Detection | 84.1 img/sec |
| 2.87 Gpixels/sec | Horizon Detection | 3.07 Gpixels/sec |
| 3.76 Gpixels/sec | Edge Detection | 3.91 Gpixels/sec |
| 2.89 Gpixels/sec | Gaussian Blur | 3.85 Gpixels/sec |
| 0.81 Gpixels/sec | Feature Matching | 1.1 Gpixels/sec |
| 270.9 Gpixels/sec | Stereo Matching | 325.9 Gpixels/sec |
| 9339.7 FPS | Particle Physics | 11008.8 FPS |
| OpenCL | API | OpenCL |
| 12503 | G3D Mark Score | 14319 |
| 918 | G2D Mark | 630 |
| 90 FPS | DirectX 11 | 129 FPS |
| 59 FPS | DirectX 12 | 61 FPS |
| 6438 Ops/s | GPU Compute | 5914 Ops/s |
| 11848 | GB6 ML Single Precision | 15570 |
| 21368 | GB6 ML Half Precision | 27160 |
| 7905 | GB6 ML Quantized | 12230 |
| 4958 | Image Classification (SP) | 6233 |
| 12747 | Image Segmentation (HP) | 16395 |
| 9816 | Image Super Resolution (Q) | 18240 |
| 22467 | Face Detection (HP) | 30758 |
| 41362 | Pose Estimation (Q) | 61300 |
| 2563 | Text Classification (SP) | 3124 |
| 3721 | Machine Translation (HP) | 4357 |
| 5929 | Object Detection (SP) | 7812 |
| 18361 | Depth Estimation (Q) | 29192 |
| 84665 | Style Transfer (SP) | 117759 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 57 GPixel/s | Pixel Fill Rate | 51-76 GPixel/s |
| 142 GTexel/s | Texture Fill Rate | 128-190 GTexel/s |
| 9.1 TFLOPS | FLOPS (FP32) | 8.2-12.1 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 8 GB | Memory Size | 6 GB |
| 1750 MHz | Memory Clock | 2000 MHz |
| 14000 Mbps | Effective Memory Speed | 16000 Mbps |
| 128-bit | Bus | 96-bit |
| No | ECC | No |
| 224 GB/s | Memory Bandwidth | 192 GB/s |
| PCIe 4.0 x8 | Interface | PCIe 4.0 x16 |
| 130 W | TGP | 35-140 W (configurable) |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 5 nm |
| 276 mm² | Die Size | 156 mm² |
| 12 billion | Transistor Count | 18.9 billion |
| 43.48 MTr/mm² | Transistor Density | 121.15 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.6 | CUDA | 8.9 |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | DLSS 3 |
| 1.4a | DisplayPort | 1.4a |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| January 27, 2022 | Released | February 22, 2023 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Gaming |
| Entry-level | Segment | Entry-level |
| Ampere | Architecture | Ada Lovelace |
| GA106 | GPU Codename | AD107 (GN21-X2) |
| -Intel Core i5 12400or above | Recommended CPU | -Intel Core Ultra 7 155Hor above |
| 6352 | Steel Nomad Lite Score | 7949 |
| 6235 | Time Spy | 8225 |
| 30139 | Solar Bay | 35461 |
| 3545 | Port Royal | 4631 |
| 15968 | Fire Strike | 21904 |
| 12803 | Wild Life Extreme | 15531 |
| 74799 | Night Raid | 85428 |
| 1552 MHz | Base Clock | 1140-2070 MHz (configurable) |
| 1777 MHz | Boost Clock | 1605-2370 MHz (configurable) |
| 2560 | Shading Units | 2560 |
| 80 | Texture Mapping Units (TMUs) | 80 |
| 32 | Render Output Units (ROPs) | 32 |
| 20 | Compute Units (Pipelines) | 20 |
| 80 | Tensor Cores | 80 |
| 20 | Ray-tracing Cores | 20 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 2MB shared | L2 Cache | 12MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |