Quick/Comparisons
60 | Gaming | 6 |
64 | Workstation | 12 |
46 | AI/ML |
Compare with something else?
Compare GeForce RTX 3080 with another →4 |
35 | Energy Efficiency | 84 |
59 | Quick Comparison Final Score | 20 |
| GA102 | Graphics Processor | Raven |
| 10240 | Cores | 512 |
| 320 / 112 | TMUs / ROPs | 32 / 8 |
| 5565.9 | Blender GPU | — |
| 128 | Forza Horizon 5 | 13 |
| 174 | The Witcher 3 | 10 |
| 205 | Counter-Strike 2 | 15 |
| 139 | Far Cry 6 | 16 |
| 108 | Hogwarts Legacy | - |
| 161 | Call of Duty | 16 |
| 96 | Ghost of Tsushima | 9 |
| 128 | Cyberpunk 2077 | 9 |
| 219 | Tomb Raider | 16 |
| 151 | Average FPS | 12 |
| 180 | 1080p High | 16 |
| 151 | 1080p Ultra | 12 |
| 117 | 1440p Ultra | 9 |
| 69 | 4K Ultra | - |
| Low | Margin of Error | Low |
| 196756 | GB6 Compute Score | 10472 |
| 239.8 img/sec | Background Blur | 21.2 img/sec |
| 156.8 img/sec | Face Detection | 10.9 img/sec |
| 10.8 Gpixels/sec | Horizon Detection | 0.34 Gpixels/sec |
| 14.8 Gpixels/sec | Edge Detection | 0.65 Gpixels/sec |
| 11.9 Gpixels/sec | Gaussian Blur | 0.51 Gpixels/sec |
| 1.53 Gpixels/sec | Feature Matching | 0.13 Gpixels/sec |
| 889.6 Gpixels/sec | Stereo Matching | 34.9 Gpixels/sec |
| 26054.6 FPS | Particle Physics | 1322.9 FPS |
| OpenCL | API | OpenCL |
| 26777 | G3D Mark Score | 1566 |
| 1095 | G2D Mark | 400 |
| 222 FPS | DirectX 11 | 10 FPS |
| 109 FPS | DirectX 12 | 8 FPS |
| 15025 Ops/s | GPU Compute | 863 Ops/s |
| 25945 | GB6 ML Single Precision | 2257 |
| 38430 | GB6 ML Half Precision | 2817 |
| 15331 | GB6 ML Quantized | 1643 |
| 9747 | Image Classification (SP) | 924 |
| 26654 | Image Segmentation (HP) | 1148 |
| 21475 | Image Super Resolution (Q) | 2211 |
| 49813 | Face Detection (HP) | 2819 |
| 135370 | Pose Estimation (Q) | 7106 |
| 2796 | Text Classification (SP) | 832 |
| 4757 | Machine Translation (HP) | 1021 |
| 12924 | Object Detection (SP) | 1040 |
| 36756 | Depth Estimation (Q) | 3885 |
| 274032 | Style Transfer (SP) | 14495 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 186 GPixel/s | Pixel Fill Rate | 9-17 GPixel/s |
| 533 GTexel/s | Texture Fill Rate | 35-67 GTexel/s |
| 34.1 TFLOPS | FLOPS (FP32) | 1.1-2.2 TFLOPS |
| GDDR6X | Memory Type | System Shared |
| 12 GB | Memory Size | System RAM Dependent |
| 1188 MHz | Memory Clock | System RAM Dependent |
| 19000 Mbps | Effective Memory Speed | System RAM Dependent |
| 384-bit | Bus | — |
| No | ECC | No |
| 912 GB/s | Memory Bandwidth | — |
| PCIe 4.0 x16 | Interface | IGP |
| 350 W | TGP | 15 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 7 nm |
| 628 mm² | Die Size | 210 mm² |
| 28 billion | Transistor Count | 4.94 billion |
| 44.59 MTr/mm² | Transistor Density | 23.52 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2 |
| 8.6 | CUDA | — |
| Yes | Ray Tracing | No |
| DLSS 2 | DLSS | No |
| 1.4a | DisplayPort | — |
| Nvidia | Vendor | Amd |
| Discrete | Build | Integrated |
| January 3, 2021 | Released | February 13, 2018 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Generic |
| High-end | Segment | Entry-level |
| Ampere | Architecture | GCN 5.0 |
| GA102 | GPU Codename | Raven |
| -Intel Core i9 12900Kor above | Recommended CPU | -AMD Ryzen 7 4800Uor above |
| 21408 | Steel Nomad Lite Score | 1260 |
| 19632 | Time Spy | 1410 |
| 96597 | Solar Bay | — |
| 13265 | Port Royal | — |
| 47867 | Fire Strike | 4499 |
| 42716 | Wild Life Extreme | 2532 |
| 144955 | Night Raid | 17993 |
| 1365 MHz | Base Clock | 300 MHz |
| 1665 MHz | Boost Clock | 1100-2100 MHz (configurable) |
| 10240 | Shading Units | 512 |
| 320 | Texture Mapping Units (TMUs) | 32 |
| 112 | Render Output Units (ROPs) | 8 |
| 80 | Compute Units (Pipelines) | 8 |
| 320 | Tensor Cores | No |
| 80 | Ray-tracing Cores | No |
| 128KB per cluster | L1 Cache | — |
| 6MB shared | L2 Cache | — |
| 2 IPC | Instructions Per Cycle | 2 IPC |