Quick/Comparisons
42 | Gaming | 57 |
48 | Workstation | 55 |
37 | AI/ML |
Compare with something else?
Compare GeForce RTX 3080 with another →48 |
37 | Energy Efficiency | 45 |
42 | Quick Comparison Final Score | 53 |
| GA103 (GN20-E8) | Graphics Processor | AD104 (GN21-X9) |
| 7424 | Cores | 7424 |
| 232 / 96 | TMUs / ROPs | 232 / 80 |
| 3502.46 | Blender GPU | 5463.87 |
| 92 | Forza Horizon 5 | 120 |
| 136 | The Witcher 3 | 163 |
| 181 | Counter-Strike 2 | 209 |
| 111 | Far Cry 6 | 139 |
| 79 | Hogwarts Legacy | 98 |
| 106 | Call of Duty | 153 |
| 72 | Ghost of Tsushima | 92 |
| 86 | Cyberpunk 2077 | 122 |
| 143 | Tomb Raider | 208 |
| 112 | Average FPS | 145 |
| 135 | 1080p High | 171 |
| 112 | 1080p Ultra | 145 |
| 83 | 1440p Ultra | 112 |
| 47 | 4K Ultra | 66 |
| Low | Margin of Error | Low |
| 133659 | GB6 Compute Score | 170767 |
| 200.9 img/sec | Background Blur | 328.9 img/sec |
| 131.2 img/sec | Face Detection | 192.3 img/sec |
| 6.49 Gpixels/sec | Horizon Detection | 6.51 Gpixels/sec |
| 8.9 Gpixels/sec | Edge Detection | 9.11 Gpixels/sec |
| 6.34 Gpixels/sec | Gaussian Blur | 10.1 Gpixels/sec |
| 1.2 Gpixels/sec | Feature Matching | 1.56 Gpixels/sec |
| 571 Gpixels/sec | Stereo Matching | 660.7 Gpixels/sec |
| 17240.3 FPS | Particle Physics | 20765.5 FPS |
| OpenCL | API | OpenCL |
| 18948 | G3D Mark Score | 24711 |
| 817 | G2D Mark | 922 |
| 161 FPS | DirectX 11 | 241 FPS |
| 87 FPS | DirectX 12 | 96 FPS |
| 8302 Ops/s | GPU Compute | 11008 Ops/s |
| 18415 | GB6 ML Single Precision | 26637 |
| 29828 | GB6 ML Half Precision | 39211 |
| 11425 | GB6 ML Quantized | 19995 |
| 6797 | Image Classification (SP) | 10385 |
| 20808 | Image Segmentation (HP) | 25896 |
| 15194 | Image Super Resolution (Q) | 27816 |
| 37084 | Face Detection (HP) | 51976 |
| 82268 | Pose Estimation (Q) | 142459 |
| 2863 | Text Classification (SP) | 3405 |
| 4135 | Machine Translation (HP) | 4860 |
| 8559 | Object Detection (SP) | 13308 |
| 28330 | Depth Estimation (Q) | 49203 |
| 184129 | Style Transfer (SP) | 262792 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 108-153 GPixel/s | Pixel Fill Rate | 133-182 GPixel/s |
| 261-369 GTexel/s | Texture Fill Rate | 386-529 GTexel/s |
| 16.7-23.6 TFLOPS | FLOPS (FP32) | 24.7-33.9 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 16 GB | Memory Size | 12 GB |
| 2000 MHz | Memory Clock | 2250 MHz |
| 16000 Mbps | Effective Memory Speed | 18000 Mbps |
| 256-bit | Bus | 192-bit |
| No | ECC | No |
| 512 GB/s | Memory Bandwidth | 432 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 80-175 W (configurable) | TGP | 60-175 W (configurable) |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 5 nm |
| 496 mm² | Die Size | 295 mm² |
| 22 billion | Transistor Count | 35.8 billion |
| 44.35 MTr/mm² | Transistor Density | 121.36 MTr/mm² |
| 12 | DirectX | 12 |
| 1.2 | 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 1, 2022 | Released | February 8, 2023 |
| Laptop | Case | Laptop |
| Gaming | Purpose | Gaming |
| High-end | Segment | High-end |
| Ampere | Architecture | Ada Lovelace |
| GA103 (GN20-E8) | GPU Codename | AD104 (GN21-X9) |
| -Intel Core i9 12900Hor above | Recommended CPU | -Intel Core Ultra 9 185Hor above |
| 12913 | Steel Nomad Lite Score | 19263 |
| 12864 | Time Spy | 18832 |
| 56305 | Solar Bay | 85404 |
| 8092 | Port Royal | 11837 |
| 30858 | Fire Strike | 41182 |
| 26307 | Wild Life Extreme | 36965 |
| 98582 | Night Raid | 147212 |
| 585-1230 MHz (configurable) | Base Clock | 1290-1860 MHz (configurable) |
| 1125-1590 MHz (configurable) | Boost Clock | 1665-2280 MHz (configurable) |
| 7424 | Shading Units | 7424 |
| 232 | Texture Mapping Units (TMUs) | 232 |
| 96 | Render Output Units (ROPs) | 80 |
| 58 | Compute Units (Pipelines) | 58 |
| 232 | Tensor Cores | 232 |
| 58 | Ray-tracing Cores | 58 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 4MB shared | L2 Cache | 48MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |