Quick/Comparisons
92 | Gaming | 9 |
91 | Workstation | 12 |
76 | AI/ML |
Compare with something else?
Compare GeForce RTX 4090 with another →15 |
43 | Energy Efficiency | 93 |
89 | Quick Comparison Final Score | 24 |
| AD102 | Graphics Processor | Phoenix |
| 16384 | Cores | 512 |
| 512 / 176 | TMUs / ROPs | 32 / 16 |
| 11688.48 | Blender GPU | 258.41 |
| 203 | Forza Horizon 5 | 20 |
| 270 | The Witcher 3 | 20 |
| 245 | Counter-Strike 2 | 26 |
| 183 | Far Cry 6 | 24 |
| 165 | Hogwarts Legacy | 11 |
| 269 | Call of Duty | 26 |
| 150 | Ghost of Tsushima | 12 |
| 204 | Cyberpunk 2077 | 16 |
| 283 | Tomb Raider | 21 |
| 219 | Average FPS | 20 |
| 256 | 1080p High | 25 |
| 219 | 1080p Ultra | 20 |
| 176 | 1440p Ultra | 13 |
| 107 | 4K Ultra | - |
| Low | Margin of Error | Low |
| 2.38 Gpixels/sec | Feature Matching | 0.29 Gpixels/sec |
| 1950 Gpixels/sec | Stereo Matching | 85.4 Gpixels/sec |
| 48090.4 FPS | Particle Physics | 2844.5 FPS |
| OpenCL | API | OpenCL |
| 316301 | GB6 Compute Score | 22589 |
| 318.9 img/sec | Background Blur | 57.8 img/sec |
| 222.8 img/sec | Face Detection | 28.7 img/sec |
| 14.2 Gpixels/sec | Horizon Detection | 0.8 Gpixels/sec |
| 21 Gpixels/sec | Edge Detection | 1 Gpixels/sec |
| 23.9 Gpixels/sec | Gaussian Blur | 0.82 Gpixels/sec |
| 38071 | G3D Mark Score | 5374 |
| 1303 | G2D Mark | 894 |
| 324 FPS | DirectX 11 | 53 FPS |
| 151 FPS | DirectX 12 | 25 FPS |
| 26193 Ops/s | GPU Compute | 2877 Ops/s |
| 42962 | GB6 ML Single Precision | 8097 |
| 59636 | GB6 ML Half Precision | 11246 |
| 31670 | GB6 ML Quantized | 6004 |
| 15600 | Image Classification (SP) | 3449 |
| 36657 | Image Segmentation (HP) | 4098 |
| 48114 | Image Super Resolution (Q) | 7937 |
| 78111 | Face Detection (HP) | 12168 |
| 289639 | Pose Estimation (Q) | 29915 |
| 4074 | Text Classification (SP) | 2571 |
| 6349 | Machine Translation (HP) | 3852 |
| 21186 | Object Detection (SP) | 2687 |
| 75599 | Depth Estimation (Q) | 14163 |
| 601428 | Style Transfer (SP) | 59966 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 444 GPixel/s | Pixel Fill Rate | 42-45 GPixel/s |
| 1290 GTexel/s | Texture Fill Rate | 83-90 GTexel/s |
| 82.6 TFLOPS | FLOPS (FP32) | 2.7-2.9 TFLOPS |
| GDDR6X | Memory Type | System Shared |
| 24 GB | Memory Size | — |
| 2625 MHz | Memory Clock | — |
| 21000 Mbps | Effective Memory Speed | — |
| 384-bit | Bus | — |
| No | ECC | — |
| 1010 GB/s | Memory Bandwidth | — |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x8 |
| 450 W | TGP | 15 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 4 nm |
| 609 mm² | Die Size | 178 mm² |
| 76.3 billion | Transistor Count | 25.4 billion |
| 125.29 MTr/mm² | Transistor Density | 142.7 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.1 |
| 8.9 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | No |
| 1.4a | DisplayPort | 2.1 |
| Nvidia | Vendor | Amd |
| Discrete | Build | Integrated |
| September 20, 2022 | Released | January 4, 2023 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Generic |
| High-end | Segment | Entry-level |
| Ada Lovelace | Architecture | RDNA 3.0 |
| AD102 | GPU Codename | Phoenix |
| -Intel Core i9 14900Kor above | Recommended CPU | -AMD Ryzen 5 7640Uor above |
| 42169 | Steel Nomad Lite Score | 2061 |
| 36311 | Time Spy | 2373 |
| 187386 | Solar Bay | 9378 |
| 26128 | Port Royal | 975 |
| 72387 | Fire Strike | 7087 |
| 85230 | Wild Life Extreme | 4630 |
| 195880 | Night Raid | 29647 |
| 2235 MHz | Base Clock | 800 MHz |
| 2520 MHz | Boost Clock | 2600-2800 MHz (configurable) |
| 16384 | Shading Units | 512 |
| 512 | Texture Mapping Units (TMUs) | 32 |
| 176 | Render Output Units (ROPs) | 16 |
| 128 | Compute Units (Pipelines) | 8 |
| 512 | Tensor Cores | — |
| 128 | Ray-tracing Cores | 6 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 72MB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |