Quick/Comparisons
51 | Gaming | 5 |
57 | Workstation | 13 |
42 | AI/ML |
Compare with something else?
Compare Nvidia RTX 4000 with another →6 |
45 | Energy Efficiency | 75 |
51 | Quick Comparison Final Score | 19 |
| AD104 | Graphics Processor | — |
| 7424 | Cores | 448 |
| 232 / 80 | TMUs / ROPs | 28 / 8 |
| 5009.03 | Blender GPU | — |
| 107 | Forza Horizon 5 | 10 |
| 152 | The Witcher 3 | — |
| 200 | Counter-Strike 2 | 14 |
| 126 | Far Cry 6 | 11 |
| 89 | Hogwarts Legacy | - |
| 132 | Call of Duty | 13 |
| 82 | Ghost of Tsushima | - |
| 109 | Cyberpunk 2077 | - |
| 175 | Tomb Raider | 11 |
| 130 | Average FPS | 10 |
| 157 | 1080p High | 13 |
| 130 | 1080p Ultra | 10 |
| 99 | 1440p Ultra | - |
| 58 | 4K Ultra | - |
| High | Margin of Error | Low |
| 139462 | GB6 Compute Score | 14408 |
| 187.1 img/sec | Background Blur | 33.7 img/sec |
| 139.5 img/sec | Face Detection | 15.4 img/sec |
| 5.71 Gpixels/sec | Horizon Detection | 0.35 Gpixels/sec |
| 8.53 Gpixels/sec | Edge Detection | 0.76 Gpixels/sec |
| 8.38 Gpixels/sec | Gaussian Blur | 0.69 Gpixels/sec |
| 1.55 Gpixels/sec | Feature Matching | 0.21 Gpixels/sec |
| 577.7 Gpixels/sec | Stereo Matching | 54.7 Gpixels/sec |
| 16849 FPS | Particle Physics | 1957.6 FPS |
| OpenCL | API | OpenCL |
| 22027 | G3D Mark Score | 2250 |
| 721 | G2D Mark | 568 |
| 215 FPS | DirectX 11 | 16 FPS |
| 96 FPS | DirectX 12 | 11 FPS |
| 8635 Ops/s | GPU Compute | 1215 Ops/s |
| 22488 | GB6 ML Single Precision | 3157 |
| 35441 | GB6 ML Half Precision | 4191 |
| 16689 | GB6 ML Quantized | 2365 |
| 8658 | Image Classification (SP) | 1323 |
| 17759 | Image Segmentation (HP) | 1675 |
| 28281 | Image Super Resolution (Q) | 2919 |
| 37971 | Face Detection (HP) | 4386 |
| 113009 | Pose Estimation (Q) | 9348 |
| 3535 | Text Classification (SP) | 1024 |
| 4038 | Machine Translation (HP) | 1888 |
| 11288 | Object Detection (SP) | 1452 |
| 38963 | Depth Estimation (Q) | 5272 |
| 236141 | Style Transfer (SP) | 19585 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 117-181 GPixel/s | Pixel Fill Rate | 12-16 GPixel/s |
| 340-524 GTexel/s | Texture Fill Rate | 42-56 GTexel/s |
| 21.8-33.6 TFLOPS | FLOPS (FP32) | 1.3-1.8 TFLOPS |
| GDDR6 | Memory Type | System Shared |
| 12 GB | Memory Size | System RAM Dependent |
| 2250 MHz | Memory Clock | System RAM Dependent |
| 18000 Mbps | Effective Memory Speed | System RAM Dependent |
| 192-bit | Bus | — |
| Yes | ECC | No |
| 432 GB/s | Memory Bandwidth | — |
| PCIe 4.0 x16 | Interface | IGP |
| 60-175 W (configurable) | TGP | 10-45 W (configurable) |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 7 nm |
| 294 mm² | Die Size | — |
| 35.8 billion | Transistor Count | — |
| 121.77 MTr/mm² | Transistor Density | — |
| 12 | DirectX | 12.1 |
| 1.3 | Vulkan | — |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.2 |
| 8.9 | CUDA | — |
| Yes | Ray Tracing | No |
| DLSS 3 | DLSS | No |
| 1.4a | DisplayPort | — |
| Nvidia | Vendor | Amd |
| Discrete | Build | Integrated |
| March 21, 2023 | Released | January 7, 2020 |
| Laptop | Case | Laptop |
| Professional | Purpose | Generic |
| High-end | Segment | Entry-level |
| Ada Lovelace | Architecture | Raven |
| AD104 | GPU Codename | — |
| -Intel Core Ultra 9 185Hor above | Recommended CPU | -AMD Ryzen 5 5500Uor above |
| 16926 | Steel Nomad Lite Score | 943 |
| 15516 | Time Spy | 1068 |
| 68617 | Solar Bay | — |
| 10118 | Port Royal | — |
| 38840 | Fire Strike | 3396 |
| — | Wild Life Extreme | 2015 |
| 123923 | Night Raid | 13093 |
| — | Base Clock | 300 MHz |
| 1465-2260 MHz (configurable) | Boost Clock | 1500-2000 MHz (configurable) |
| 7424 | Shading Units | 448 |
| 232 | Texture Mapping Units (TMUs) | 28 |
| 80 | Render Output Units (ROPs) | 8 |
| 58 | Compute Units (Pipelines) | 7 |
| 232 | Tensor Cores | No |
| 58 | Ray-tracing Cores | No |
| 128KB per cluster | L1 Cache | — |
| 48MB shared | L2 Cache | — |
| 2 IPC | Instructions Per Cycle | 2 IPC |