Quick/Comparisons
A mid-range Nvidia GeForce RTX 4090 Laptop graphics card with capable gaming performance (63/100). Suited for 1080p gaming and everyday tasks.
| Gaming | 63 /100 |
| Workstation | 63 /100 |
| AI/ML | 55 /100 |
| Energy Efficiency | 45 |
Looking for more comparisons?
Compare GeForce RTX →| Quick Comparison Final Score | 59 /100 |
| Graphics Processor | AD103 (GN21-X11) |
| Cores | 9728 |
| TMUs / ROPs | 304 / 112 |
| Blender GPU | 7059.06 |
| Forza Horizon 5 | 131 |
| The Witcher 3 | 182 |
| Counter-Strike 2 | 208 |
| Far Cry 6 | 147 |
| Hogwarts Legacy | 108 |
| Call of Duty | 164 |
| Ghost of Tsushima | 100 |
| Cyberpunk 2077 | 137 |
| Tomb Raider | 231 |
| Average FPS | 156 |
| 1080p High | 187 |
| 1080p Ultra | 156 |
| 1440p Ultra | 123 |
| 4K Ultra | 72 |
| Margin of Error | Low |
| GB6 Compute Score | 203263 |
| Background Blur | 264 img/sec |
| Face Detection | 134.3 img/sec |
| Horizon Detection | 9.27 Gpixels/sec |
| Edge Detection | 12.7 Gpixels/sec |
| Gaussian Blur | 13.2 Gpixels/sec |
| Feature Matching | 1.92 Gpixels/sec |
| Stereo Matching | 1060 Gpixels/sec |
| Particle Physics | 29225.1 FPS |
| API | OpenCL |
| G3D Mark Score | 27060 |
| G2D Mark | 988 |
| DirectX 11 | 258 FPS |
| DirectX 12 | 106 FPS |
| GPU Compute | 12206 Ops/s |
| GB6 ML Single Precision | 29362 |
| GB6 ML Half Precision | 42611 |
| GB6 ML Quantized | 22212 |
| Image Classification (SP) | 11346 |
| Image Segmentation (HP) | 24625 |
| Image Super Resolution (Q) | 36874 |
| Face Detection (HP) | 50088 |
| Pose Estimation (Q) | 174844 |
| Text Classification (SP) | 3226 |
| Machine Translation (HP) | 4905 |
| Object Detection (SP) | 14667 |
| Depth Estimation (Q) | 53868 |
| Style Transfer (SP) | 340056 |
| Framework | ONNX |
| Backend | DirectML |
| Pixel Fill Rate | 163-228 GPixel/s |
| Texture Fill Rate | 442-620 GTexel/s |
| FLOPS (FP32) | 28.3-39.7 TFLOPS |
| Memory Type | GDDR6 |
| Memory Size | 16 GB |
| Memory Clock | 2000 MHz |
| Effective Memory Speed | 16000 Mbps |
| Bus | 256-bit |
| ECC | No |
| Memory Bandwidth | 512 GB/s |
| Interface | PCIe 4.0 x16 |
| TGP | 80-175 W (configurable) |
| Manufacturing | TSMC |
| Fabrication Process | 5 nm |
| Die Size | 379 mm² |
| Transistor Count | 45.9 billion |
| Transistor Density | 121.11 MTr/mm² |
| DirectX | 12 |
| Vulkan | 1.3 |
| OpenGL | 4.6 |
| OpenCL | 3 |
| CUDA | 8.9 |
| Ray Tracing | Yes |
| DLSS | DLSS 3 |
| DisplayPort | 1.4a |
| Vendor | Nvidia |
| Build | Discrete |
| Released | February 8, 2023 |
| Case | Laptop |
| Purpose | Gaming |
| Segment | High-end |
| Architecture | Ada Lovelace |
| GPU Codename | AD103 (GN21-X11) |
| Recommended CPU | -Intel Core Ultra 9 185Hor above |
| Steel Nomad Lite Score | 22501 |
| Time Spy | 21401 |
| Solar Bay | 96130 |
| Port Royal | 13733 |
| Fire Strike | 45218 |
| Wild Life Extreme | 42491 |
| Night Raid | 147892 |
| Base Clock | 930-1620 MHz (configurable) |
| Boost Clock | 1455-2040 MHz (configurable) |
| Shading Units | 9728 |
| Texture Mapping Units (TMUs) | 304 |
| Render Output Units (ROPs) | 112 |
| Compute Units (Pipelines) | 76 |
| Tensor Cores | 304 |
| Ray-tracing Cores | 76 |
| L1 Cache | 128KB per cluster |
| L2 Cache | 64MB shared |
| Instructions Per Cycle | 2 IPC |
84/100