Quick/Comparisons
66 | Gaming | 9 |
73 | Workstation | 12 |
52 | AI/ML |
Compare with something else?
Compare GeForce RTX 3090 with another →15 |
37 | Energy Efficiency | 93 |
66 | Quick Comparison Final Score | 24 |
| GA102 | Graphics Processor | Phoenix |
| 10752 | Cores | 512 |
| 336 / 112 | TMUs / ROPs | 32 / 16 |
| 6057.83 | Blender GPU | 258.41 |
| 141 | Forza Horizon 5 | 20 |
| 182 | The Witcher 3 | 20 |
| 210 | Counter-Strike 2 | 26 |
| 147 | Far Cry 6 | 24 |
| 112 | Hogwarts Legacy | 11 |
| 173 | Call of Duty | 26 |
| 108 | Ghost of Tsushima | 12 |
| 143 | Cyberpunk 2077 | 16 |
| 243 | Tomb Raider | 21 |
| 162 | Average FPS | 20 |
| 128 | 1440p Ultra | 13 |
| 75 | 4K Ultra | - |
| Low | Margin of Error | Low |
| 193 | 1080p High | 25 |
| 162 | 1080p Ultra | 20 |
| 232083 | GB6 Compute Score | 22589 |
| 362.7 img/sec | Background Blur | 57.8 img/sec |
| 227.9 img/sec | Face Detection | 28.7 img/sec |
| 11.3 Gpixels/sec | Horizon Detection | 0.8 Gpixels/sec |
| 15.9 Gpixels/sec | Edge Detection | 1 Gpixels/sec |
| 12.1 Gpixels/sec | Gaussian Blur | 0.82 Gpixels/sec |
| 1.56 Gpixels/sec | Feature Matching | 0.29 Gpixels/sec |
| 1080 Gpixels/sec | Stereo Matching | 85.4 Gpixels/sec |
| 31024.6 FPS | Particle Physics | 2844.5 FPS |
| OpenCL | API | OpenCL |
| 29288 | G3D Mark Score | 5374 |
| 1226 | G2D Mark | 894 |
| 238 FPS | DirectX 11 | 53 FPS |
| 122 FPS | DirectX 12 | 25 FPS |
| 16569 Ops/s | GPU Compute | 2877 Ops/s |
| 26701 | GB6 ML Single Precision | 8097 |
| 38573 | GB6 ML Half Precision | 11246 |
| 15995 | GB6 ML Quantized | 6004 |
| 10246 | Image Classification (SP) | 3449 |
| 22830 | Image Segmentation (HP) | 4098 |
| 23532 | Image Super Resolution (Q) | 7937 |
| 48657 | Face Detection (HP) | 12168 |
| 147241 | Pose Estimation (Q) | 29915 |
| 2917 | Text Classification (SP) | 2571 |
| 4427 | Machine Translation (HP) | 3852 |
| 13323 | Object Detection (SP) | 2687 |
| 40055 | Depth Estimation (Q) | 14163 |
| 298683 | Style Transfer (SP) | 59966 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 208 GPixel/s | Pixel Fill Rate | 42-45 GPixel/s |
| 625 GTexel/s | Texture Fill Rate | 83-90 GTexel/s |
| 40 TFLOPS | FLOPS (FP32) | 2.7-2.9 TFLOPS |
| GDDR6X | Memory Type | System Shared |
| 24 GB | Memory Size | — |
| 1313 MHz | Memory Clock | — |
| 21000 Mbps | Effective Memory Speed | — |
| 384-bit | Bus | — |
| No | ECC | — |
| 1008 GB/s | Memory Bandwidth | — |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x8 |
| 450 W | TGP | 15 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 4 nm |
| 628 mm² | Die Size | 178 mm² |
| 28 billion | Transistor Count | 25.4 billion |
| 44.59 MTr/mm² | Transistor Density | 142.7 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.1 |
| 8.6 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | No |
| 1.4a | DisplayPort | 2.1 |
| Nvidia | Vendor | Amd |
| Discrete | Build | Integrated |
| March 29, 2022 | Released | January 4, 2023 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Generic |
| High-end | Segment | Entry-level |
| Ampere | Architecture | RDNA 3.0 |
| GA102 | GPU Codename | Phoenix |
| -Intel Core i9 12900Kor above | Recommended CPU | -AMD Ryzen 5 7640Uor above |
| 24378 | Steel Nomad Lite Score | 2061 |
| 21854 | Time Spy | 2373 |
| 109167 | Solar Bay | 9378 |
| 14888 | Port Royal | 975 |
| 51943 | Fire Strike | 7087 |
| 48864 | Wild Life Extreme | 4630 |
| 156500 | Night Raid | 29647 |
| 1560 MHz | Base Clock | 800 MHz |
| 1860 MHz | Boost Clock | 2600-2800 MHz (configurable) |
| 10752 | Shading Units | 512 |
| 336 | Texture Mapping Units (TMUs) | 32 |
| 112 | Render Output Units (ROPs) | 16 |
| 84 | Compute Units (Pipelines) | 8 |
| 336 | Tensor Cores | — |
| 84 | Ray-tracing Cores | 6 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 6MB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |