Quick/Comparisons
The GeForce RTX 4080 SUPER's 78 score is significantly higher than the Nvidia T550 Laptop's 10, thanks to its superior performance in gaming workloads.
The GeForce RTX 4080 SUPER's 76 score in content creation workloads is higher than the Nvidia T550 Laptop's 21, driven by its larger 16GB memory pool.
The GeForce RTX 4080 SUPER's 66 score in AI/ML workloads is higher than the Nvidia T550 Laptop's 9, despite not being the clear winner in this category.
78 | Gaming | 10 |
76 | Workstation | 21 |
66 | AI/ML |
Compare with something else?
Compare GeForce RTX 4080 with another →9 |
46 | Energy Efficiency | 73 |
75 | Quick Comparison Final Score | 23 |
| AD103 | Graphics Processor | TU117 |
| 10240 | Cores | 1024 |
| 320 / 112 | TMUs / ROPs | 64 / 32 |
| 8420.48 | Blender GPU | 142.83 |
| 171 | Forza Horizon 5 | — |
| 210 | The Witcher 3 | — |
| 223 | Counter-Strike 2 | — |
| 166 | Far Cry 6 | — |
| 137 | Hogwarts Legacy | — |
| 216 | Call of Duty | — |
| 122 | Ghost of Tsushima | — |
| 177 | Cyberpunk 2077 | — |
| 265 | Tomb Raider | — |
| 187 | Average FPS | — |
| 221 | 1080p High | — |
| 187 | 1080p Ultra | — |
| 149 | 1440p Ultra | — |
| 87 | 4K Ultra | — |
| Low | Margin of Error | — |
| 250691 | GB6 Compute Score | 35511 |
| 279.5 img/sec | Background Blur | 61.9 img/sec |
| 220.1 img/sec | Face Detection | 28.6 img/sec |
| 11 Gpixels/sec | Horizon Detection | 1.2 Gpixels/sec |
| 15.2 Gpixels/sec | Edge Detection | 1.67 Gpixels/sec |
| 16.2 Gpixels/sec | Gaussian Blur | 1.9 Gpixels/sec |
| 2.08 Gpixels/sec | Feature Matching | 0.5 Gpixels/sec |
| 1370 Gpixels/sec | Stereo Matching | 174.2 Gpixels/sec |
| 37199.7 FPS | Particle Physics | 4869.6 FPS |
| OpenCL | API | OpenCL |
| 34252 | G3D Mark Score | 4752 |
| 1278 | G2D Mark | 401 |
| 300 FPS | DirectX 11 | 35 FPS |
| 134 FPS | DirectX 12 | 30 FPS |
| 19639 Ops/s | GPU Compute | 2256 Ops/s |
| 36883 | GB6 ML Single Precision | 4976 |
| 54286 | GB6 ML Half Precision | 4623 |
| 28474 | GB6 ML Quantized | 3647 |
| 12453 | Image Classification (SP) | 2063 |
| 34467 | Image Segmentation (HP) | 3228 |
| 43534 | Image Super Resolution (Q) | 4420 |
| 70300 | Face Detection (HP) | 5053 |
| 242331 | Pose Estimation (Q) | 14361 |
| 3767 | Text Classification (SP) | 1515 |
| 6614 | Machine Translation (HP) | 2212 |
| 18301 | Object Detection (SP) | 2394 |
| 66477 | Depth Estimation (Q) | 8645 |
| 417906 | Style Transfer (SP) | 33780 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 286 GPixel/s | Pixel Fill Rate | 53 GPixel/s |
| 816 GTexel/s | Texture Fill Rate | 107 GTexel/s |
| 52.2 TFLOPS | FLOPS (FP32) | 3.4 TFLOPS |
| GDDR6X | Memory Type | GDDR6 |
| 16 GB | Memory Size | 4 GB |
| 1438 MHz | Memory Clock | 1500 MHz |
| 23000 Mbps | Effective Memory Speed | 12000 Mbps |
| 256-bit | Bus | 64-bit |
| No | ECC | No |
| 736.3 GB/s | Memory Bandwidth | 96 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 3.0 x16 |
| 320 W | TGP | 30 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 12 nm |
| 379 mm² | Die Size | 200 mm² |
| 45 billion | Transistor Count | 4.7 billion |
| 118.73 MTr/mm² | Transistor Density | 23.5 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.9 | CUDA | 7.5 |
| Yes | Ray Tracing | No |
| DLSS 3 | DLSS | No |
| 1.4a | DisplayPort | 1.4 |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| January 31, 2024 | Released | May 1, 2022 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Professional |
| High-end | Segment | Entry-level |
| Ada Lovelace | Architecture | Turing |
| AD103 | GPU Codename | TU117 |
| -Intel Core i7 14700Kor above | Recommended CPU | -Intel Core i7 1265Uor above |
| 30711 | Steel Nomad Lite Score | 1984 |
| 28279 | Time Spy | 2273 |
| 138998 | Solar Bay | — |
| 18348 | Port Royal | — |
| 63714 | Fire Strike | 5443 |
| 60015 | Wild Life Extreme | 4481 |
| 180412 | Night Raid | 28589 |
| 2295 MHz | Base Clock | 1065 MHz |
| 2550 MHz | Boost Clock | 1665 MHz |
| 10240 | Shading Units | 1024 |
| 320 | Texture Mapping Units (TMUs) | 64 |
| 112 | Render Output Units (ROPs) | 32 |
| 80 | Compute Units (Pipelines) | 16 |
| 320 | Tensor Cores | — |
| 80 | Ray-tracing Cores | — |
| 128KB per cluster | L1 Cache | 64KB per cluster |
| 64MB shared | L2 Cache | 1MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |