Quick/Comparisons
The GeForce RTX 5090 dominates in gaming performance, outscoring the GeForce RTX 3050 Ti Laptop by 80 points. Its exceptional gaming capabilities make it the clear winner.
The GeForce RTX 5090 excels in workstation performance, outscoring the GeForce RTX 3050 Ti Laptop by 67 points. Its strong performance in content creation tasks makes it the winner.
The GeForce RTX 5090 leads in AI/ML compute, outscoring the GeForce RTX 3050 Ti Laptop by 75 points. Its dedicated AI tensor cores and strong performance make it the winner.
20 | Gaming | 100 |
26 | Workstation | 93 |
20 | AI/ML | 95 |
59 | Energy Efficiency | 49 |
28 | Quick Comparison Final Score | 96 |
| GA107 (GN20-P1) | Graphics Processor | GB202-300 |
| 2560 | Cores | 21760 |
| 80 / 48 | TMUs / ROPs | 680 / 176 |
| 1323.51 | Blender GPU | 15010.34 |
| 50 | Forza Horizon 5 | 244 |
| 59 | The Witcher 3 | 323 |
| 89 | Counter-Strike 2 | 265 |
| 58 | Far Cry 6 | 202 |
| 35 | Hogwarts Legacy | 207 |
| 43 | Call of Duty | 327 |
| 32 | Ghost of Tsushima | 179 |
| 39 | Cyberpunk 2077 | 229 |
| 69 | Tomb Raider | 299 |
| 53 | Average FPS | 253 |
| 67 | 1080p High | 294 |
| 53 | 1080p Ultra | 253 |
| 37 | 1440p Ultra | 207 |
| 19 | 4K Ultra | 124 |
| Low | Margin of Error | Low |
| 65121 | GB6 Compute Score | 419309 |
| 107.9 img/sec | Background Blur | 333 img/sec |
| 62.6 img/sec | Face Detection | 298.5 img/sec |
| 2.75 Gpixels/sec | Horizon Detection | 21.7 Gpixels/sec |
| 3.61 Gpixels/sec | Edge Detection | 33.7 Gpixels/sec |
| 2.89 Gpixels/sec | Gaussian Blur | 34.5 Gpixels/sec |
| 0.81 Gpixels/sec | Feature Matching | 2.53 Gpixels/sec |
| 275.9 Gpixels/sec | Stereo Matching | 2850 Gpixels/sec |
| 8303.3 FPS | Particle Physics | 59146.9 FPS |
| OpenCL | API | OpenCL |
| 10086 | G3D Mark Score | 38955 |
| 498 | G2D Mark | 1413 |
| 75 FPS | DirectX 11 | 341 FPS |
| 49 FPS | DirectX 12 | 178 FPS |
| 4322 Ops/s | GPU Compute | 24632 Ops/s |
| 11209 | GB6 ML Single Precision | 53861 |
| 20383 | GB6 ML Half Precision | 81080 |
| 7783 | GB6 ML Quantized | 38938 |
| 4716 | Image Classification (SP) | 18948 |
| 12032 | Image Segmentation (HP) | 55109 |
| 9965 | Image Super Resolution (Q) | 58478 |
| 20723 | Face Detection (HP) | 114685 |
| 41186 | Pose Estimation (Q) | 344729 |
| 2133 | Text Classification (SP) | 4464 |
| 3449 | Machine Translation (HP) | 8184 |
| 5585 | Object Detection (SP) | 27802 |
| 19119 | Depth Estimation (Q) | 85572 |
| 86331 | Style Transfer (SP) | 687080 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 50-81 GPixel/s | Pixel Fill Rate | 424 GPixel/s |
| 83-136 GTexel/s | Texture Fill Rate | 1637 GTexel/s |
| 5.3-8.7 TFLOPS | FLOPS (FP32) | 104.8 TFLOPS |
| GDDR6 | Memory Type | GDDR7 |
| 4 GB | Memory Size | 32 GB |
| 1500 MHz | Memory Clock | 1750 MHz |
| 12000 Mbps | Effective Memory Speed | 28000 Mbps |
| 128-bit | Bus | 512-bit |
| No | ECC | No |
| 192 GB/s | Memory Bandwidth | 1792 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 35-95 W (configurable) | TGP | 575 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 4 nm |
| 276 mm² | Die Size | 750 mm² |
| 12 billion | Transistor Count | 92.2 billion |
| 43.48 MTr/mm² | Transistor Density | 122.93 MTr/mm² |
| 12 | DirectX | 12.2 |
| 1.2 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.6 | CUDA | 12 |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | DLSS 4 |
| 1.4a | DisplayPort | 2.1b |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| January 1, 2021 | Released | January 7, 2025 |
| Laptop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Ampere | Architecture | Blackwell 2.0 |
| GA107 (GN20-P1) | GPU Codename | GB202-300 |
| -Intel Core i7 12700Hor above | Recommended CPU | -Intel Core Ultra 9 285Kor above |
| 5684 | Steel Nomad Lite Score | 52638 |
| 5383 | Time Spy | 46941 |
| 24147 | Solar Bay | 235224 |
| 570 | Port Royal | 38113 |
| 13499 | Fire Strike | 87433 |
| 10255 | Wild Life Extreme | 108736 |
| 56622 | Night Raid | 207058 |
| 735-1463 MHz (configurable) | Base Clock | 2017 MHz |
| 1035-1695 MHz (configurable) | Boost Clock | 2407 MHz |
| 2560 | Shading Units | 21760 |
| 80 | Texture Mapping Units (TMUs) | 680 |
| 48 | Render Output Units (ROPs) | 176 |
| 20 | Compute Units (Pipelines) | 170 |
| 80 | Tensor Cores | 680 |
| 20 | Ray-tracing Cores | 170 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 2MB shared | L2 Cache | 96MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |
Compare with something else?
Compare GeForce RTX 3050 with another →