Quick/Comparisons
The GeForce RTX 4080 SUPER outperforms the GeForce RTX 3060 Ti in gaming with a score of 78, thanks to its superior performance and efficiency. Its 40-point lead is a significant advantage in this use case.
Both the GeForce RTX 3060 Ti and GeForce RTX 4080 SUPER deliver strong performance in content creation, with the GeForce RTX 3060 Ti edging out slightly with a score of 42. However, the difference is only 4 points, making this a close tie.
The GeForce RTX 4080 SUPER takes the lead in AI and machine learning with a score of 66, thanks to its dedicated AI tensor cores and improved architecture. Its 35-point lead is a significant advantage in this use case.
38 | Gaming | 78 |
42 | Workstation | 76 |
31 | AI/ML | 66 |
35 | Energy Efficiency | 46 |
38 | Quick Comparison Final Score | 75 |
| GA104 | Graphics Processor | AD103 |
| 4864 | Cores | 10240 |
| 152 / 80 | TMUs / ROPs | 320 / 112 |
| 2992.04 | Blender GPU | 8420.48 |
| 87 | Forza Horizon 5 | 171 |
| 124 | The Witcher 3 | 210 |
| 167 | Counter-Strike 2 | 223 |
| 106 | Far Cry 6 | 166 |
| 71 | Hogwarts Legacy | 137 |
| 100 | Call of Duty | 216 |
| 68 | Ghost of Tsushima | 122 |
| 82 | Cyberpunk 2077 | 177 |
| 128 | Tomb Raider | 265 |
| 104 | Average FPS | 187 |
| 127 | 1080p High | 221 |
| 104 | 1080p Ultra | 187 |
| 77 | 1440p Ultra | 149 |
| 43 | 4K Ultra | 87 |
| Low | Margin of Error | Low |
| 115518 | GB6 Compute Score | 250691 |
| 185.1 img/sec | Background Blur | 279.5 img/sec |
| 108.4 img/sec | Face Detection | 220.1 img/sec |
| 5.35 Gpixels/sec | Horizon Detection | 11 Gpixels/sec |
| 7.17 Gpixels/sec | Edge Detection | 15.2 Gpixels/sec |
| 5.54 Gpixels/sec | Gaussian Blur | 16.2 Gpixels/sec |
| 1.15 Gpixels/sec | Feature Matching | 2.08 Gpixels/sec |
| 478 Gpixels/sec | Stereo Matching | 1370 Gpixels/sec |
| 15183 FPS | Particle Physics | 37199.7 FPS |
| OpenCL | API | OpenCL |
| 20272 | G3D Mark Score | 34252 |
| 990 | G2D Mark | 1278 |
| 164 FPS | DirectX 11 | 300 FPS |
| 78 FPS | DirectX 12 | 134 FPS |
| 9905 Ops/s | GPU Compute | 19639 Ops/s |
| 17811 | GB6 ML Single Precision | 36883 |
| 29969 | GB6 ML Half Precision | 54286 |
| 11367 | GB6 ML Quantized | 28474 |
| 7384 | Image Classification (SP) | 12453 |
| 19320 | Image Segmentation (HP) | 34467 |
| 16108 | Image Super Resolution (Q) | 43534 |
| 34615 | Face Detection (HP) | 70300 |
| 74811 | Pose Estimation (Q) | 242331 |
| 2638 | Text Classification (SP) | 3767 |
| 4136 | Machine Translation (HP) | 6614 |
| 8718 | Object Detection (SP) | 18301 |
| 25946 | Depth Estimation (Q) | 66477 |
| 153275 | Style Transfer (SP) | 417906 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 133 GPixel/s | Pixel Fill Rate | 286 GPixel/s |
| 253 GTexel/s | Texture Fill Rate | 816 GTexel/s |
| 16.2 TFLOPS | FLOPS (FP32) | 52.2 TFLOPS |
| GDDR6 | Memory Type | GDDR6X |
| 8 GB | Memory Size | 16 GB |
| 1750 MHz | Memory Clock | 1438 MHz |
| 14000 Mbps | Effective Memory Speed | 23000 Mbps |
| 256-bit | Bus | 256-bit |
| No | ECC | No |
| 448 GB/s | Memory Bandwidth | 736.3 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 200 W | TGP | 320 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 5 nm |
| 392 mm² | Die Size | 379 mm² |
| 17 billion | Transistor Count | 45 billion |
| 43.37 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 |
| December 2, 2020 | Released | January 31, 2024 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Ampere | Architecture | Ada Lovelace |
| GA104 | GPU Codename | AD103 |
| -Intel Core i5 12600Kor above | Recommended CPU | -Intel Core i7 14700Kor above |
| 11978 | Steel Nomad Lite Score | 30711 |
| 11707 | Time Spy | 28279 |
| 54374 | Solar Bay | 138998 |
| 6971 | Port Royal | 18348 |
| 29568 | Fire Strike | 63714 |
| 24415 | Wild Life Extreme | 60015 |
| 109854 | Night Raid | 180412 |
| 2 IPC | Instructions Per Cycle | 2 IPC |
| 1410 MHz | Base Clock | 2295 MHz |
| 1665 MHz | Boost Clock | 2550 MHz |
| 4864 | Shading Units | 10240 |
| 152 | Texture Mapping Units (TMUs) | 320 |
| 80 | Render Output Units (ROPs) | 112 |
| 38 | Compute Units (Pipelines) | 80 |
| 152 | Tensor Cores | 320 |
| 38 | Ray-tracing Cores | 80 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 4MB shared | L2 Cache | 64MB shared |
Compare with something else?
Compare GeForce RTX 3060 with another →