Quick/Comparisons
9 | Gaming | 57 |
16 | Workstation | 49 |
11 | AI/ML |
Compare with something else?
Compare Radeon 680M with another →56 |
88 | Energy Efficiency | 36 |
24 | Quick Comparison Final Score | 54 |
| Rembrandt | Graphics Processor | Navi 21 |
| 768 | Cores | 5120 |
| 48 / 32 | TMUs / ROPs | 320 / 128 |
| 238.39 | Blender GPU | 2484.02 |
| 19 | Forza Horizon 5 | 139 |
| 18 | The Witcher 3 | 185 |
| 26 | Counter-Strike 2 | 208 |
| 26 | Far Cry 6 | 152 |
| 13 | Hogwarts Legacy | 103 |
| 24 | Call of Duty | 179 |
| 13 | Ghost of Tsushima | 112 |
| 16 | Cyberpunk 2077 | 127 |
| 25 | Tomb Raider | 208 |
| 20 | Average FPS | 157 |
| 26 | 1080p High | 188 |
| 20 | 1080p Ultra | 157 |
| 13 | 1440p Ultra | 121 |
| - | 4K Ultra | 71 |
| Low | Margin of Error | Low |
| 28675 | GB6 Compute Score | 124121 |
| 73.8 img/sec | Background Blur | 231.6 img/sec |
| 30.1 img/sec | Face Detection | 152.5 img/sec |
| 0.93 Gpixels/sec | Horizon Detection | 4.26 Gpixels/sec |
| 1.17 Gpixels/sec | Edge Detection | 5.75 Gpixels/sec |
| 1.17 Gpixels/sec | Gaussian Blur | 5.87 Gpixels/sec |
| 0.37 Gpixels/sec | Feature Matching | 1.12 Gpixels/sec |
| 105 Gpixels/sec | Stereo Matching | 618.7 Gpixels/sec |
| 4640 FPS | Particle Physics | 17973.9 FPS |
| OpenCL | API | OpenCL |
| 4325 | G3D Mark Score | 26671 |
| 734 | G2D Mark | 1062 |
| 39 FPS | DirectX 11 | 279 FPS |
| 20 FPS | DirectX 12 | 111 FPS |
| 1980 Ops/s | GPU Compute | 14356 Ops/s |
| 5834 | GB6 ML Single Precision | 33651 |
| 8202 | GB6 ML Half Precision | 36994 |
| 4169 | GB6 ML Quantized | 24546 |
| 1976 | Image Classification (SP) | 13868 |
| 3138 | Image Segmentation (HP) | 17467 |
| 6693 | Image Super Resolution (Q) | 32396 |
| 7213 | Face Detection (HP) | 47808 |
| 21849 | Pose Estimation (Q) | 178873 |
| 1497 | Text Classification (SP) | 4358 |
| 2579 | Machine Translation (HP) | 7910 |
| 2154 | Object Detection (SP) | 15394 |
| 9635 | Depth Estimation (Q) | 57351 |
| 56269 | Style Transfer (SP) | 372610 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 70-77 GPixel/s | Pixel Fill Rate | 288 GPixel/s |
| 106-115 GTexel/s | Texture Fill Rate | 720 GTexel/s |
| 3.4-3.7 TFLOPS | FLOPS (FP32) | 23 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 16 GB |
| — | Memory Clock | 2000 MHz |
| — | Effective Memory Speed | 16000 Mbps |
| — | Bus | 256-bit |
| No | ECC | No |
| — | Memory Bandwidth | 512 GB/s |
| PCIe 4.0 x8 | Interface | PCIe 4.0 x16 |
| 15 W | TGP | 300 W |
| TSMC | Manufacturing | TSMC |
| 6 nm | Fabrication Process | 7 nm |
| 208 mm² | Die Size | 520 mm² |
| 13.1 billion | Transistor Count | 26 billion |
| 62.98 MTr/mm² | Transistor Density | 50 MTr/mm² |
| 12 | DirectX | 12 |
| 1.2 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 2 | OpenCL | 2.1 |
| — | CUDA | — |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| — | DisplayPort | 1.4a |
| Amd | Vendor | Amd |
| Integrated | Build | Discrete |
| January 1, 2022 | Released | December 8, 2020 |
| Laptop | Case | Desktop |
| Generic | Purpose | Gaming |
| Entry-level | Segment | High-end |
| RDNA 2.0 | Architecture | RDNA 2.0 |
| Rembrandt | GPU Codename | Navi 21 |
| -AMD Ryzen 7 6800Hor above | Recommended CPU | -Intel Core i7 12700Kor above |
| 2141 | Steel Nomad Lite Score | 20125 |
| 2328 | Time Spy | 20809 |
| 9717 | Solar Bay | 86980 |
| 1010 | Port Royal | 10539 |
| 6897 | Fire Strike | 57319 |
| 4167 | Wild Life Extreme | 39341 |
| 27577 | Night Raid | 159586 |
| 400 MHz | Base Clock | 1825 MHz |
| 2200-2400 MHz (configurable) | Boost Clock | 2250 MHz |
| 768 | Shading Units | 5120 |
| 48 | Texture Mapping Units (TMUs) | 320 |
| 32 | Render Output Units (ROPs) | 128 |
| 12 | Compute Units (Pipelines) | — |
| — | Tensor Cores | — |
| 12 | Ray-tracing Cores | 80 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 2MB shared | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |