Quick/Comparisons
31 | Gaming | 46 |
37 | Workstation | 49 |
29 | AI/ML |
Compare with something else?
Compare GeForce RTX 3060 with another →39 |
37 | Energy Efficiency | 35 |
33 | Quick Comparison Final Score | 46 |
| GA106 | Graphics Processor | GA104 |
| 3584 | Cores | 6144 |
| 112 / 48 | TMUs / ROPs | 192 / 96 |
| 2303.56 | Blender GPU | 3746.83 |
| 71 | Forza Horizon 5 | 104 |
| 100 | The Witcher 3 | 143 |
| 133 | Counter-Strike 2 | 198 |
| 83 | Far Cry 6 | 125 |
| 56 | Hogwarts Legacy | 85 |
| 73 | Call of Duty | 120 |
| 49 | Ghost of Tsushima | 81 |
| 62 | Cyberpunk 2077 | 102 |
| 97 | Tomb Raider | 161 |
| 80 | Average FPS | 124 |
| 100 | 1080p High | 150 |
| 80 | 1080p Ultra | 124 |
| 59 | 1440p Ultra | 93 |
| 32 | 4K Ultra | 54 |
| Low | Margin of Error | Low |
| 91603 | GB6 Compute Score | 136096 |
| 149.6 img/sec | Background Blur | 197 img/sec |
| 91 img/sec | Face Detection | 128.1 img/sec |
| 4.07 Gpixels/sec | Horizon Detection | 6.56 Gpixels/sec |
| 5.58 Gpixels/sec | Edge Detection | 8.81 Gpixels/sec |
| 3.92 Gpixels/sec | Gaussian Blur | 6.47 Gpixels/sec |
| 1.01 Gpixels/sec | Feature Matching | 1.26 Gpixels/sec |
| 372.9 Gpixels/sec | Stereo Matching | 609.3 Gpixels/sec |
| 12226.3 FPS | Particle Physics | 18205 FPS |
| OpenCL | API | OpenCL |
| 16744 | G3D Mark Score | 23260 |
| 966 | G2D Mark | 1061 |
| 128 FPS | DirectX 11 | 192 FPS |
| 69 FPS | DirectX 12 | 91 FPS |
| 8266 Ops/s | GPU Compute | 11452 Ops/s |
| 15123 | GB6 ML Single Precision | 22516 |
| 25942 | GB6 ML Half Precision | 36377 |
| 10356 | GB6 ML Quantized | 14215 |
| 6645 | Image Classification (SP) | 8615 |
| 15338 | Image Segmentation (HP) | 24659 |
| 14283 | Image Super Resolution (Q) | 20631 |
| 28862 | Face Detection (HP) | 43834 |
| 61313 | Pose Estimation (Q) | 98347 |
| 2463 | Text Classification (SP) | 3162 |
| 3793 | Machine Translation (HP) | 4911 |
| 7712 | Object Detection (SP) | 11498 |
| 25461 | Depth Estimation (Q) | 34719 |
| 123233 | Style Transfer (SP) | 196833 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 85 GPixel/s | Pixel Fill Rate | 170 GPixel/s |
| 199 GTexel/s | Texture Fill Rate | 340 GTexel/s |
| 12.7 TFLOPS | FLOPS (FP32) | 21.7 TFLOPS |
| GDDR6 | Memory Type | GDDR6X |
| 12 GB | Memory Size | 8 GB |
| 1875 MHz | Memory Clock | 1188 MHz |
| 15000 Mbps | Effective Memory Speed | 19000 Mbps |
| 192-bit | Bus | 256-bit |
| No | ECC | No |
| 360 GB/s | Memory Bandwidth | 608.3 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 170 W | TGP | 290 W |
| Samsung | Manufacturing | Samsung |
| 8 nm | Fabrication Process | 8 nm |
| 276 mm² | Die Size | 392 mm² |
| 12 billion | Transistor Count | 17 billion |
| 43.48 MTr/mm² | Transistor Density | 43.37 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 |
| DLSS 2 | DLSS | DLSS 2 |
| 1.4a | DisplayPort | 1.4a |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| January 12, 2021 | Released | June 10, 2021 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Ampere | Architecture | Ampere |
| GA106 | GPU Codename | GA104 |
| -Intel Core i5 12600Kor above | Recommended CPU | -Intel Core i7 12700Kor above |
| 9002 | Steel Nomad Lite Score | 15402 |
| 8754 | Time Spy | 14880 |
| 41139 | Solar Bay | 69900 |
| 5145 | Port Royal | 8884 |
| 22306 | Fire Strike | 37238 |
| 17726 | Wild Life Extreme | 31911 |
| 97592 | Night Raid | 134012 |
| 1320 MHz | Base Clock | 1575 MHz |
| 1777 MHz | Boost Clock | 1770 MHz |
| 3584 | Shading Units | 6144 |
| 112 | Texture Mapping Units (TMUs) | 192 |
| 48 | Render Output Units (ROPs) | 96 |
| 28 | Compute Units (Pipelines) | 48 |
| 112 | Tensor Cores | 192 |
| 28 | Ray-tracing Cores | 48 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 3MB shared | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |