Quick/Comparisons
66 | Gaming | 78 |
73 | Workstation | 76 |
52 | AI/ML |
Compare with something else?
Compare GeForce RTX 3090 with another →66 |
37 | Energy Efficiency | 46 |
66 | Quick Comparison Final Score | 75 |
| GA102 | Graphics Processor | AD103 |
| 10752 | Cores | 10240 |
| 336 / 112 | TMUs / ROPs | 320 / 112 |
| 6057.83 | Blender GPU | 8420.48 |
| 141 | Forza Horizon 5 | 171 |
| 182 | The Witcher 3 | 210 |
| 210 | Counter-Strike 2 | 223 |
| 147 | Far Cry 6 | 166 |
| 112 | Hogwarts Legacy | 137 |
| 173 | Call of Duty | 216 |
| 108 | Ghost of Tsushima | 122 |
| 143 | Cyberpunk 2077 | 177 |
| 243 | Tomb Raider | 265 |
| 162 | Average FPS | 187 |
| 128 | 1440p Ultra | 149 |
| 75 | 4K Ultra | 87 |
| Low | Margin of Error | Low |
| 193 | 1080p High | 221 |
| 162 | 1080p Ultra | 187 |
| 232083 | GB6 Compute Score | 250691 |
| 362.7 img/sec | Background Blur | 279.5 img/sec |
| 227.9 img/sec | Face Detection | 220.1 img/sec |
| 11.3 Gpixels/sec | Horizon Detection | 11 Gpixels/sec |
| 15.9 Gpixels/sec | Edge Detection | 15.2 Gpixels/sec |
| 12.1 Gpixels/sec | Gaussian Blur | 16.2 Gpixels/sec |
| 1.56 Gpixels/sec | Feature Matching | 2.08 Gpixels/sec |
| 1080 Gpixels/sec | Stereo Matching | 1370 Gpixels/sec |
| 31024.6 FPS | Particle Physics | 37199.7 FPS |
| OpenCL | API | OpenCL |
| 29288 | G3D Mark Score | 34252 |
| 1226 | G2D Mark | 1278 |
| 238 FPS | DirectX 11 | 300 FPS |
| 122 FPS | DirectX 12 | 134 FPS |
| 16569 Ops/s | GPU Compute | 19639 Ops/s |
| 26701 | GB6 ML Single Precision | 36883 |
| 38573 | GB6 ML Half Precision | 54286 |
| 15995 | GB6 ML Quantized | 28474 |
| 10246 | Image Classification (SP) | 12453 |
| 22830 | Image Segmentation (HP) | 34467 |
| 23532 | Image Super Resolution (Q) | 43534 |
| 48657 | Face Detection (HP) | 70300 |
| 147241 | Pose Estimation (Q) | 242331 |
| 2917 | Text Classification (SP) | 3767 |
| 4427 | Machine Translation (HP) | 6614 |
| 13323 | Object Detection (SP) | 18301 |
| 40055 | Depth Estimation (Q) | 66477 |
| 298683 | Style Transfer (SP) | 417906 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 208 GPixel/s | Pixel Fill Rate | 286 GPixel/s |
| 625 GTexel/s | Texture Fill Rate | 816 GTexel/s |
| 40 TFLOPS | FLOPS (FP32) | 52.2 TFLOPS |
| GDDR6X | Memory Type | GDDR6X |
| 24 GB | Memory Size | 16 GB |
| 1313 MHz | Memory Clock | 1438 MHz |
| 21000 Mbps | Effective Memory Speed | 23000 Mbps |
| 384-bit | Bus | 256-bit |
| No | ECC | No |
| 1008 GB/s | Memory Bandwidth | 736.3 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 450 W | TGP | 320 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 5 nm |
| 628 mm² | Die Size | 379 mm² |
| 28 billion | Transistor Count | 45 billion |
| 44.59 MTr/mm² | Transistor Density | 118.73 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | 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 |
| March 29, 2022 | Released | January 31, 2024 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | High-end |
| Ampere | Architecture | Ada Lovelace |
| GA102 | GPU Codename | AD103 |
| -Intel Core i9 12900Kor above | Recommended CPU | -Intel Core i7 14700Kor above |
| 24378 | Steel Nomad Lite Score | 30711 |
| 21854 | Time Spy | 28279 |
| 109167 | Solar Bay | 138998 |
| 14888 | Port Royal | 18348 |
| 51943 | Fire Strike | 63714 |
| 48864 | Wild Life Extreme | 60015 |
| 156500 | Night Raid | 180412 |
| 1560 MHz | Base Clock | 2295 MHz |
| 1860 MHz | Boost Clock | 2550 MHz |
| 10752 | Shading Units | 10240 |
| 336 | Texture Mapping Units (TMUs) | 320 |
| 112 | Render Output Units (ROPs) | 112 |
| 84 | Compute Units (Pipelines) | 80 |
| 336 | Tensor Cores | 320 |
| 84 | Ray-tracing Cores | 80 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 6MB shared | L2 Cache | 64MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |