Quick/Comparisons
60 | Gaming | 73 |
64 | Workstation | 62 |
46 | AI/ML |
Compare with something else?
Compare GeForce RTX 3080 with another →67 |
35 | Energy Efficiency | 49 |
59 | Quick Comparison Final Score | 68 |
| GA102 | Graphics Processor | Navi 48 |
| 10240 | Cores | 4096 |
| 320 / 112 | TMUs / ROPs | 256 / 128 |
| 5565.9 | Blender GPU | 3176.42 |
| 128 | Forza Horizon 5 | 169 |
| 174 | The Witcher 3 | 236 |
| 205 | Counter-Strike 2 | 224 |
| 139 | Far Cry 6 | 178 |
| 108 | Hogwarts Legacy | 132 |
| 161 | Call of Duty | 234 |
| 96 | Ghost of Tsushima | 135 |
| 128 | Cyberpunk 2077 | 166 |
| 219 | Tomb Raider | 268 |
| 151 | Average FPS | 194 |
| 180 | 1080p High | 229 |
| 151 | 1080p Ultra | 194 |
| 117 | 1440p Ultra | 150 |
| 69 | 4K Ultra | 89 |
| Low | Margin of Error | Low |
| 196756 | GB6 Compute Score | 182816 |
| 239.8 img/sec | Background Blur | 427.4 img/sec |
| 156.8 img/sec | Face Detection | 251 img/sec |
| 10.8 Gpixels/sec | Horizon Detection | 6.77 Gpixels/sec |
| 14.8 Gpixels/sec | Edge Detection | 9.42 Gpixels/sec |
| 11.9 Gpixels/sec | Gaussian Blur | 9.35 Gpixels/sec |
| 1.53 Gpixels/sec | Feature Matching | 1.42 Gpixels/sec |
| 889.6 Gpixels/sec | Stereo Matching | 651.4 Gpixels/sec |
| 26054.6 FPS | Particle Physics | 23630 FPS |
| OpenCL | API | OpenCL |
| 26777 | G3D Mark Score | 26904 |
| 1095 | G2D Mark | 1319 |
| 222 FPS | DirectX 11 | 295 FPS |
| 109 FPS | DirectX 12 | 77 FPS |
| 15025 Ops/s | GPU Compute | 15995 Ops/s |
| 25945 | GB6 ML Single Precision | 39190 |
| 38430 | GB6 ML Half Precision | 55727 |
| 15331 | GB6 ML Quantized | 33599 |
| 9747 | Image Classification (SP) | 18001 |
| 26654 | Image Segmentation (HP) | 25174 |
| 21475 | Image Super Resolution (Q) | 45714 |
| 49813 | Face Detection (HP) | 74776 |
| 135370 | Pose Estimation (Q) | 196137 |
| 2796 | Text Classification (SP) | 5163 |
| 4757 | Machine Translation (HP) | 8677 |
| 12924 | Object Detection (SP) | 19602 |
| 36756 | Depth Estimation (Q) | 74879 |
| 274032 | Style Transfer (SP) | 411302 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 186 GPixel/s | Pixel Fill Rate | 380 GPixel/s |
| 533 GTexel/s | Texture Fill Rate | 760 GTexel/s |
| 34.1 TFLOPS | FLOPS (FP32) | 48.7 TFLOPS |
| GDDR6X | Memory Type | GDDR6 |
| 12 GB | Memory Size | 16 GB |
| 1188 MHz | Memory Clock | 2518 MHz |
| 19000 Mbps | Effective Memory Speed | 20100 Mbps |
| 384-bit | Bus | 256-bit |
| No | ECC | No |
| 912 GB/s | Memory Bandwidth | 640 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 350 W | TGP | 304 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 4 nm |
| 628 mm² | Die Size | 356.5 mm² |
| 28 billion | Transistor Count | 53.9 billion |
| 44.59 MTr/mm² | Transistor Density | 151.19 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.2 |
| 8.6 | CUDA | No |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | No |
| 1.4a | DisplayPort | 2.1a |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| January 3, 2021 | Released | January 7, 2025 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | High-end |
| Ampere | Architecture | RDNA 4.0 |
| GA102 | GPU Codename | Navi 48 |
| -Intel Core i9 12900Kor above | Recommended CPU | -Intel Core Ultra 5 245Kor above |
| 21408 | Steel Nomad Lite Score | 26180 |
| 19632 | Time Spy | 29701 |
| 96597 | Solar Bay | 121225 |
| 13265 | Port Royal | 18525 |
| 47867 | Fire Strike | 66381 |
| 42716 | Wild Life Extreme | 54685 |
| 144955 | Night Raid | 195698 |
| 1365 MHz | Base Clock | 1660 MHz |
| 1665 MHz | Boost Clock | 2970 MHz |
| 10240 | Shading Units | 4096 |
| 320 | Texture Mapping Units (TMUs) | 256 |
| 112 | Render Output Units (ROPs) | 128 |
| 80 | Compute Units (Pipelines) | 64 |
| 320 | Tensor Cores | 128 |
| 80 | Ray-tracing Cores | 64 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 6MB shared | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 4 IPC |