Quick/Comparisons
61 | Gaming | 16 |
69 | Workstation | 25 |
53 | AI/ML |
Compare with something else?
Compare GeForce RTX 3090 with another →22 |
35 | Energy Efficiency | 100 |
62 | Quick Comparison Final Score | 33 |
| GA102 | Graphics Processor | Panther Lake |
| 10496 | Cores | 1536 |
| 328 / 112 | TMUs / ROPs | 48 / 24 |
| 5800.59 | Blender GPU | 1268.79 |
| 125 | Forza Horizon 5 | 38 |
| 175 | The Witcher 3 | 55 |
| 212 | Counter-Strike 2 | 60 |
| 142 | Far Cry 6 | 49 |
| 106 | Hogwarts Legacy | 21 |
| 159 | Call of Duty | 32 |
| 97 | Ghost of Tsushima | 25 |
| 132 | Cyberpunk 2077 | 33 |
| 226 | Tomb Raider | 52 |
| 153 | Average FPS | 41 |
| 179 | 1080p High | 52 |
| 153 | 1080p Ultra | 41 |
| 119 | 1440p Ultra | 29 |
| 70 | 4K Ultra | 14 |
| Low | Margin of Error | Low |
| 211669 | GB6 Compute Score | 64546 |
| 320.8 img/sec | Background Blur | — |
| 207.2 img/sec | Face Detection | — |
| 10.1 Gpixels/sec | Horizon Detection | — |
| 14.4 Gpixels/sec | Edge Detection | — |
| 11.5 Gpixels/sec | Gaussian Blur | — |
| 1.48 Gpixels/sec | Feature Matching | — |
| 984.2 Gpixels/sec | Stereo Matching | — |
| 27764.9 FPS | Particle Physics | — |
| OpenCL | API | — |
| 26551 | G3D Mark Score | 9029 |
| 1069 | G2D Mark | 786 |
| 218 FPS | DirectX 11 | 70 FPS |
| 110 FPS | DirectX 12 | 50 FPS |
| 15117 Ops/s | GPU Compute | 3702 Ops/s |
| 27237 | GB6 ML Single Precision | 11871 |
| 41500 | GB6 ML Half Precision | 20120 |
| 15927 | GB6 ML Quantized | 7879 |
| 10665 | Image Classification (SP) | 5487 |
| 28963 | Image Segmentation (HP) | 16864 |
| 22697 | Image Super Resolution (Q) | 8788 |
| 55232 | Face Detection (HP) | 25029 |
| 128155 | Pose Estimation (Q) | 18852 |
| 3152 | Text Classification (SP) | 2115 |
| 5088 | Machine Translation (HP) | 4266 |
| 13891 | Object Detection (SP) | 7449 |
| 37984 | Depth Estimation (Q) | 22207 |
| 272000 | Style Transfer (SP) | 95039 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 190 GPixel/s | Pixel Fill Rate | 60 GPixel/s |
| 556 GTexel/s | Texture Fill Rate | 120 GTexel/s |
| 35.6 TFLOPS | FLOPS (FP32) | 7.7 TFLOPS |
| GDDR6X | Memory Type | System Shared |
| 24 GB | Memory Size | — |
| 1219 MHz | Memory Clock | — |
| 19500 Mbps | Effective Memory Speed | — |
| 384-bit | Bus | — |
| No | ECC | No |
| 936.2 GB/s | Memory Bandwidth | — |
| PCIe 4.0 x16 | Interface | PCIe 5.0 |
| 350 W | TGP | — |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 3 nm |
| 628 mm² | Die Size | — |
| 28 billion | Transistor Count | — |
| 44.59 MTr/mm² | Transistor Density | — |
| 12 | DirectX | 12.1 |
| 1.3 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.6 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | No |
| 1.4a | DisplayPort | 2.1 |
| Nvidia | Vendor | Intel |
| Discrete | Build | Integrated |
| September 24, 2020 | Released | January 5, 2026 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Gaming |
| High-end | Segment | Mid-range |
| Ampere | Architecture | Celestial |
| GA102 | GPU Codename | Panther Lake |
| -Intel Core i9 12900Kor above | Recommended CPU | -Intel Core Ultra X9 388Hor above |
| 21851 | Steel Nomad Lite Score | 4242 |
| 19896 | Time Spy | 6595 |
| 97375 | Solar Bay | 26399 |
| 13631 | Port Royal | 3886 |
| 47515 | Fire Strike | 16770 |
| 43673 | Wild Life Extreme | 11707 |
| 139755 | Night Raid | 68498 |
| 1395 MHz | Base Clock | 300 MHz |
| 1695 MHz | Boost Clock | 2500 MHz |
| 10496 | Shading Units | 1536 |
| 328 | Texture Mapping Units (TMUs) | 48 |
| 112 | Render Output Units (ROPs) | 24 |
| 82 | Compute Units (Pipelines) | 12 |
| 328 | Tensor Cores | — |
| 82 | Ray-tracing Cores | — |
| 128KB per cluster | L1 Cache | 64KB per cluster |
| 6MB shared | L2 Cache | 16MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |