Quick/Comparisons
The GeForce RTX 5090 dominates in gaming performance, outscoring the GeForce RTX 3060 Laptop by 71 points. Its exceptional gaming capabilities make it the clear winner in this use case.
The GeForce RTX 5090's strong workstation performance gives it a 58-point lead over the GeForce RTX 3060 Laptop, making it the winner in this use case.
The GeForce RTX 5090's capable AI/ML compute capabilities result in a 69-point lead over the GeForce RTX 3060 Laptop, making it the winner in this use case.
29 | Gaming | 100 |
35 | Workstation | 93 |
26 | AI/ML | 95 |
46 | Energy Efficiency | 49 |
33 | Quick Comparison Final Score | 96 |
| GA106 (GN20-E3) | Graphics Processor | GB202-300 |
| 3840 | Cores | 21760 |
| 120 / 48 | TMUs / ROPs | 680 / 176 |
| 2342.93 | Blender GPU | 15010.34 |
| 63 | Forza Horizon 5 | 244 |
| 93 | The Witcher 3 | 323 |
| 126 | Counter-Strike 2 | 265 |
| 82 | Far Cry 6 | 202 |
| 49 | Hogwarts Legacy | 207 |
| 68 | Call of Duty | 327 |
| 44 | Ghost of Tsushima | 179 |
| 57 | Cyberpunk 2077 | 229 |
| 92 | Tomb Raider | 299 |
| 75 | Average FPS | 253 |
| 94 | 1080p High | 294 |
| 75 | 1080p Ultra | 253 |
| 55 | 1440p Ultra | 207 |
| 30 | 4K Ultra | 124 |
| Low | Margin of Error | Low |
| 94459 | GB6 Compute Score | 419309 |
| 153.5 img/sec | Background Blur | 333 img/sec |
| 90 img/sec | Face Detection | 298.5 img/sec |
| 4.1 Gpixels/sec | Horizon Detection | 21.7 Gpixels/sec |
| 5.77 Gpixels/sec | Edge Detection | 33.7 Gpixels/sec |
| 4.35 Gpixels/sec | Gaussian Blur | 34.5 Gpixels/sec |
| 1.01 Gpixels/sec | Feature Matching | 2.53 Gpixels/sec |
| 385.9 Gpixels/sec | Stereo Matching | 2850 Gpixels/sec |
| 12770 FPS | Particle Physics | 59146.9 FPS |
| OpenCL | API | OpenCL |
| 13143 | G3D Mark Score | 38955 |
| 585 | G2D Mark | 1413 |
| 109 FPS | DirectX 11 | 341 FPS |
| 58 FPS | DirectX 12 | 178 FPS |
| 5649 Ops/s | GPU Compute | 24632 Ops/s |
| 14362 | GB6 ML Single Precision | 53861 |
| 25888 | GB6 ML Half Precision | 81080 |
| 9677 | GB6 ML Quantized | 38938 |
| 6257 | Image Classification (SP) | 18948 |
| 16583 | Image Segmentation (HP) | 55109 |
| 12510 | Image Super Resolution (Q) | 58478 |
| 29528 | Face Detection (HP) | 114685 |
| 51349 | Pose Estimation (Q) | 344729 |
| 2685 | Text Classification (SP) | 4464 |
| 3939 | Machine Translation (HP) | 8184 |
| 7206 | Object Detection (SP) | 27802 |
| 24343 | Depth Estimation (Q) | 85572 |
| 113151 | Style Transfer (SP) | 687080 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 62-82 GPixel/s | Pixel Fill Rate | 424 GPixel/s |
| 154-204 GTexel/s | Texture Fill Rate | 1637 GTexel/s |
| 9.9-13.1 TFLOPS | FLOPS (FP32) | 104.8 TFLOPS |
| GDDR6 | Memory Type | GDDR7 |
| 6 GB | Memory Size | 32 GB |
| 1750 MHz | Memory Clock | 1750 MHz |
| 12000-14000 Mbps | Effective Memory Speed | 28000 Mbps |
| 192-bit | Bus | 512-bit |
| No | ECC | No |
| 336 GB/s | Memory Bandwidth | 1792 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 60-140 W (configurable) | TGP | 575 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 4 nm |
| 300 mm² | Die Size | 750 mm² |
| 13.25 billion | Transistor Count | 92.2 billion |
| 44.17 MTr/mm² | Transistor Density | 122.93 MTr/mm² |
| 12 | DirectX | 12.2 |
| 1.2 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 2 | 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 12, 2021 | Released | January 7, 2025 |
| Laptop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Ampere | Architecture | Blackwell 2.0 |
| GA106 (GN20-E3) | GPU Codename | GB202-300 |
| -Intel Core i5 12500Hor above | Recommended CPU | -Intel Core Ultra 9 285Kor above |
| 8163 | Steel Nomad Lite Score | 52638 |
| 8319 | Time Spy | 46941 |
| 36350 | Solar Bay | 235224 |
| 4709 | Port Royal | 38113 |
| 20179 | Fire Strike | 87433 |
| 16476 | Wild Life Extreme | 108736 |
| 72260 | Night Raid | 207058 |
| 816-1387 MHz (configurable) | Base Clock | 2017 MHz |
| 1283-1703 MHz (configurable) | Boost Clock | 2407 MHz |
| 3840 | Shading Units | 21760 |
| 120 | Texture Mapping Units (TMUs) | 680 |
| 48 | Render Output Units (ROPs) | 176 |
| 30 | Compute Units (Pipelines) | 170 |
| 120 | Tensor Cores | 680 |
| 30 | Ray-tracing Cores | 170 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 3MB shared | L2 Cache | 96MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |
Compare with something else?
Compare GeForce RTX 3060 with another →