Quick/Comparisons
6 | Gaming | 64 |
9 | Workstation | 60 |
9 | AI/ML |
Compare with something else?
Compare Radeon 840M with another →59 |
86 | Energy Efficiency | 44 |
20 | Quick Comparison Final Score | 62 |
| Krackan Point | Graphics Processor | Navi 31 |
| 256 | Cores | 5120 |
| 16 / 8 | TMUs / ROPs | 320 / 160 |
| 122.02 | Blender GPU | 2808.96 |
| 13 | Forza Horizon 5 | 143 |
| 12 | The Witcher 3 | 201 |
| 18 | Counter-Strike 2 | 217 |
| 16 | Far Cry 6 | 155 |
| - | Hogwarts Legacy | 109 |
| 19 | Call of Duty | 190 |
| 9 | Ghost of Tsushima | 116 |
| 9 | Cyberpunk 2077 | 138 |
| 17 | Tomb Raider | 221 |
| 13 | Average FPS | 166 |
| 18 | 1080p High | 198 |
| 13 | 1080p Ultra | 166 |
| 10 | 1440p Ultra | 128 |
| - | 4K Ultra | 74 |
| High | Margin of Error | Low |
| 14025 | GB6 Compute Score | 172224 |
| 31.8 img/sec | Background Blur | 352.4 img/sec |
| 16.8 img/sec | Face Detection | 229 img/sec |
| 0.46 Gpixels/sec | Horizon Detection | 6.42 Gpixels/sec |
| 0.58 Gpixels/sec | Edge Detection | 8.86 Gpixels/sec |
| 0.42 Gpixels/sec | Gaussian Blur | 7.89 Gpixels/sec |
| 0.18 Gpixels/sec | Feature Matching | 1.5 Gpixels/sec |
| 52.5 Gpixels/sec | Stereo Matching | 662.8 Gpixels/sec |
| 2048.6 FPS | Particle Physics | 24012 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 27367 |
| — | G2D Mark | 1207 |
| — | DirectX 11 | 300 FPS |
| — | DirectX 12 | 110 FPS |
| — | GPU Compute | 15171 Ops/s |
| 4884 | GB6 ML Single Precision | 34533 |
| 6067 | GB6 ML Half Precision | 42224 |
| 3946 | GB6 ML Quantized | 26568 |
| 1988 | Image Classification (SP) | 15095 |
| 2184 | Image Segmentation (HP) | 21206 |
| 4788 | Image Super Resolution (Q) | 33962 |
| 6857 | Face Detection (HP) | 57614 |
| 17396 | Pose Estimation (Q) | 214748 |
| 1628 | Text Classification (SP) | 4325 |
| 2053 | Machine Translation (HP) | 6598 |
| 1825 | Object Detection (SP) | 16608 |
| 8471 | Depth Estimation (Q) | 57731 |
| 31032 | Style Transfer (SP) | 394987 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 23 GPixel/s | Pixel Fill Rate | 359 GPixel/s |
| 46 GTexel/s | Texture Fill Rate | 718 GTexel/s |
| 1.5 TFLOPS | FLOPS (FP32) | 46 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 16 GB |
| — | Memory Clock | 2250 MHz |
| — | Effective Memory Speed | 18000 Mbps |
| — | Bus | 256-bit |
| — | ECC | No |
| — | Memory Bandwidth | 576 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 15-54 W (configurable) | TGP | 260 W |
| TSMC | Manufacturing | TSMC |
| 4 nm | Fabrication Process | 5 nm |
| — | Die Size | 529 mm² |
| — | Transistor Count | 57 billion |
| — | Transistor Density | 107.75 MTr/mm² |
| 12 | DirectX | 12 |
| 1.4 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 2.1 | OpenCL | 2.2 |
| No | CUDA | — |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| 2.1 | DisplayPort | 2.1 |
| Amd | Vendor | Amd |
| Integrated | Build | Discrete |
| January 7, 2025 | Released | — |
| Laptop | Case | Desktop |
| Generic | Purpose | Gaming |
| Entry-level | Segment | High-end |
| RDNA 3.5 | Architecture | RDNA 3.0 |
| Krackan Point | GPU Codename | Navi 31 |
| -AMD Ryzen AI 5 340or above | Recommended CPU | — |
| 1293 | Steel Nomad Lite Score | 21359 |
| 1700 | Time Spy | 22537 |
| 7155 | Solar Bay | 90757 |
| 577 | Port Royal | 12461 |
| 5317 | Fire Strike | 56212 |
| — | Wild Life Extreme | 41740 |
| 22727 | Night Raid | 169038 |
| 400 MHz | Base Clock | 1287 MHz |
| 2900 MHz | Boost Clock | 2245 MHz |
| 256 | Shading Units | 5120 |
| 16 | Texture Mapping Units (TMUs) | 320 |
| 8 | Render Output Units (ROPs) | 160 |
| 4 | Compute Units (Pipelines) | — |
| No | Tensor Cores | — |
| 4 | Ray-tracing Cores | 80 |
| 512KB per cluster | L1 Cache | 256KB per cluster |
| 2MB shared | L2 Cache | 6MB shared |
| 2 IPC | Instructions Per Cycle | 4 IPC |