Quick/Comparisons
7 | Gaming | 22 |
13 | Workstation | 17 |
7 | AI/ML |
Compare with something else?
Compare Radeon 660M with another →| — |
88 | Energy Efficiency | 66 |
22 | Quick Comparison Final Score | 28 |
| Rembrandt | Graphics Processor | Strix Halo |
| 384 | Cores | 1024 |
| 24 / 16 | TMUs / ROPs | 64 / 32 |
| 92.56 | Blender GPU | 376.14 |
| 9 | Ghost of Tsushima | — |
| 9 | Cyberpunk 2077 | — |
| 12 | Forza Horizon 5 | — |
| 10 | The Witcher 3 | — |
| 20 | Counter-Strike 2 | — |
| 19 | Far Cry 6 | — |
| - | Hogwarts Legacy | — |
| 15 | Call of Duty | — |
| 13 | Tomb Raider | — |
| 13 | Average FPS | — |
| 17 | 1080p High | — |
| 13 | 1080p Ultra | — |
| 9 | 1440p Ultra | — |
| - | 4K Ultra | — |
| Low | Margin of Error | — |
| 14065 | GB6 Compute Score | 35976 |
| 36 img/sec | Background Blur | 106.8 img/sec |
| 16.4 img/sec | Face Detection | 43.8 img/sec |
| 0.48 Gpixels/sec | Horizon Detection | 1.12 Gpixels/sec |
| 0.65 Gpixels/sec | Edge Detection | 1.17 Gpixels/sec |
| 0.53 Gpixels/sec | Gaussian Blur | 1.07 Gpixels/sec |
| 0.18 Gpixels/sec | Feature Matching | 0.5 Gpixels/sec |
| 46.5 Gpixels/sec | Stereo Matching | 163 Gpixels/sec |
| 2056.5 FPS | Particle Physics | 5820.4 FPS |
| OpenCL | API | OpenCL |
| 3207 | G3D Mark Score | 10431 |
| 645 | G2D Mark | 1094 |
| 54 FPS | DirectX 11 | 79 FPS |
| 27 FPS | DirectX 12 | 46 FPS |
| 1418 Ops/s | GPU Compute | 5066 Ops/s |
| 3903 | GB6 ML Single Precision | — |
| 4745 | GB6 ML Half Precision | — |
| 2977 | GB6 ML Quantized | — |
| 1465 | Image Classification (SP) | — |
| 1706 | Image Segmentation (HP) | — |
| 4266 | Image Super Resolution (Q) | — |
| 5115 | Face Detection (HP) | — |
| 12992 | Pose Estimation (Q) | — |
| 1116 | Text Classification (SP) | — |
| 1859 | Machine Translation (HP) | — |
| 1574 | Object Detection (SP) | — |
| 6538 | Depth Estimation (Q) | — |
| 26656 | Style Transfer (SP) | — |
| ONNX | Framework | — |
| DirectML | Backend | — |
| 29-30 GPixel/s | Pixel Fill Rate | 90 GPixel/s |
| 43-46 GTexel/s | Texture Fill Rate | 179 GTexel/s |
| 1.4-1.5 TFLOPS | FLOPS (FP32) | 5.7 TFLOPS |
| System Shared | Memory Type | System Shared |
| — | Memory Size | — |
| — | Memory Clock | — |
| — | Effective Memory Speed | — |
| — | Bus | — |
| No | ECC | — |
| — | Memory Bandwidth | — |
| PCIe 4.0 x8 | Interface | PCIe 5.0 x16 |
| 15 W | TGP | 45-120 W (configurable) |
| TSMC | Manufacturing | TSMC |
| 6 nm | Fabrication Process | 4 nm |
| 208 mm² | Die Size | 308 mm² |
| 13.1 billion | Transistor Count | — |
| 62.98 MTr/mm² | Transistor Density | — |
| 12 | DirectX | 12.2 |
| 1.2 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 2 | OpenCL | 2.1 |
| — | CUDA | No |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| — | DisplayPort | 2.1a |
| Amd | Vendor | Amd |
| Integrated | Build | Integrated |
| January 1, 2022 | Released | January 7, 2025 |
| Laptop | Case | Laptop |
| Generic | Purpose | Generic |
| Entry-level | Segment | Mid-range |
| RDNA 2.0 | Architecture | RDNA 3.5 |
| Rembrandt | GPU Codename | Strix Halo |
| -AMD Ryzen 5 6600Hor above | Recommended CPU | -AMD Ryzen AI Max Pro 380or above |
| 1344 | Steel Nomad Lite Score | — |
| 1515 | Time Spy | — |
| 6396 | Solar Bay | — |
| 586 | Port Royal | — |
| 4870 | Fire Strike | — |
| 2676 | Wild Life Extreme | — |
| 19259 | Night Raid | — |
| 400 MHz | Base Clock | 1295 MHz |
| 1800-1900 MHz (configurable) | Boost Clock | 2800 MHz |
| 384 | Shading Units | 1024 |
| 24 | Texture Mapping Units (TMUs) | 64 |
| 16 | Render Output Units (ROPs) | 32 |
| 6 | Compute Units (Pipelines) | 16 |
| — | Tensor Cores | No |
| 6 | Ray-tracing Cores | 16 |
| 128KB per cluster | L1 Cache | 512KB per cluster |
| 1024KB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |