Quick/Comparisons
39 | Gaming | 56 |
45 | Workstation | 67 |
35 | AI/ML |
Compare with something else?
Compare GeForce RTX 3080 with another →53 |
39 | Energy Efficiency | 49 |
40 | Quick Comparison Final Score | 57 |
| GA104 (GN20-E7) | Graphics Processor | GB203-300-A1 |
| 6144 | Cores | 7680 |
| 192 / 96 | TMUs / ROPs | 240 / 96 |
| 3258.81 | Blender GPU | 5699.95 |
| 84 | Forza Horizon 5 | 114 |
| 127 | The Witcher 3 | 157 |
| 165 | Counter-Strike 2 | 204 |
| 104 | Far Cry 6 | 132 |
| 70 | Hogwarts Legacy | 96 |
| 98 | Call of Duty | 143 |
| 66 | Ghost of Tsushima | 87 |
| 84 | Cyberpunk 2077 | 118 |
| 129 | Tomb Raider | 197 |
| 103 | Average FPS | 139 |
| 125 | 1080p High | 166 |
| 103 | 1080p Ultra | 139 |
| 76 | 1440p Ultra | 106 |
| 44 | 4K Ultra | 63 |
| Low | Margin of Error | High |
| 128827 | GB6 Compute Score | 197909 |
| 211 img/sec | Background Blur | 267.1 img/sec |
| 123.6 img/sec | Face Detection | 165.5 img/sec |
| 5.78 Gpixels/sec | Horizon Detection | 8.72 Gpixels/sec |
| 8.03 Gpixels/sec | Edge Detection | 12.3 Gpixels/sec |
| 6.25 Gpixels/sec | Gaussian Blur | 12.6 Gpixels/sec |
| 1.17 Gpixels/sec | Feature Matching | 1.83 Gpixels/sec |
| 543.2 Gpixels/sec | Stereo Matching | 942.6 Gpixels/sec |
| 17774.9 FPS | Particle Physics | 25637.1 FPS |
| OpenCL | API | OpenCL |
| 16147 | G3D Mark Score | 23401 |
| 630 | G2D Mark | 874 |
| 145 FPS | DirectX 11 | 232 FPS |
| 71 FPS | DirectX 12 | 109 FPS |
| 7126 Ops/s | GPU Compute | 8154 Ops/s |
| 18354 | GB6 ML Single Precision | 27844 |
| 29767 | GB6 ML Half Precision | 48544 |
| 11487 | GB6 ML Quantized | 21478 |
| 6773 | Image Classification (SP) | 11723 |
| 19587 | Image Segmentation (HP) | 32142 |
| 15553 | Image Super Resolution (Q) | 30735 |
| 35885 | Face Detection (HP) | 58722 |
| 82446 | Pose Estimation (Q) | 129551 |
| 2830 | Text Classification (SP) | 3539 |
| 4019 | Machine Translation (HP) | 6242 |
| 8380 | Object Detection (SP) | 14495 |
| 28597 | Depth Estimation (Q) | 50690 |
| 184715 | Style Transfer (SP) | 231154 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 120-164 GPixel/s | Pixel Fill Rate | 163-180 GPixel/s |
| 239-328 GTexel/s | Texture Fill Rate | 408-450 GTexel/s |
| 15.3-21 TFLOPS | FLOPS (FP32) | 26.1-28.8 TFLOPS |
| GDDR6 | Memory Type | GDDR7 |
| 8-16 GB | Memory Size | 16 GB |
| 1500-1750 MHz | Memory Clock | 1750 MHz |
| 12000-14000 Mbps | Effective Memory Speed | 28000 Mbps |
| 256-bit | Bus | 256-bit |
| No | ECC | No |
| 448 GB/s | Memory Bandwidth | 896 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 80-165 W (configurable) | TGP | 80-175 W (configurable) |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 5 nm |
| 392 mm² | Die Size | 378 mm² |
| 17.4 billion | Transistor Count | 45.6 billion |
| 44.39 MTr/mm² | Transistor Density | 120.63 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 | March 18, 2025 |
| Laptop | Case | Laptop |
| Gaming | Purpose | Professional |
| High-end | Segment | High-end |
| Ampere | Architecture | Blackwell 2.0 |
| GA104 (GN20-E7) | GPU Codename | GB203-300-A1 |
| -Intel Core i9 12900Hor above | Recommended CPU | -Intel Core Ultra 9 275HXor above |
| 11592 | Steel Nomad Lite Score | 22035 |
| 11991 | Time Spy | 13961 |
| 50371 | Solar Bay | — |
| 7201 | Port Royal | 12094 |
| 29067 | Fire Strike | 38402 |
| 24760 | Wild Life Extreme | — |
| 81949 | Night Raid | — |
| 780-1350 MHz (configurable) | Base Clock | 975-1702 MHz (configurable) |
| 1245-1710 MHz (configurable) | Boost Clock | 1702-1875 MHz (configurable) |
| 6144 | Shading Units | 7680 |
| 192 | Texture Mapping Units (TMUs) | 240 |
| 96 | Render Output Units (ROPs) | 96 |
| 48 | Compute Units (Pipelines) | 60 |
| 192 | Tensor Cores | 240 |
| 48 | Ray-tracing Cores | 60 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 4MB shared | L2 Cache | 64MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |