Technical Specifications
Performance
| Architecture | RDNA 2.0 |
|---|---|
| GPU Codename | Rembrandt |
| Base Clock | 400 MHz |
| Boost Clock | 2200-2400 MHz (configurable) |
| Shading Units | 768 |
| Ray-tracing Cores | 12 |
| Compute Units (Pipelines) | 12 |
Memory
| Memory Type | System Shared |
|---|
Graphics APIs
| DirectX | 12 |
|---|---|
| Vulkan | 1.2 |
| OpenGL | 4.6 |
| OpenCL | 2.0 |
| Ray Tracing | Yes |
| DLSS | No |
Technical
| TGP | 15 W |
|---|---|
| Fabrication Process | 6 nm |
| Die Size | 208 mm² |
| Transistor Count | 13.1 billion |
| Interface | PCIe 4.0 x8 |
| Manufacturing | TSMC |
| Vendor | Amd |
| Build | Integrated |
| Released | January 1, 2022 |
Other Specifications
| Forza Horizon 5 | 43201710 |
|---|---|
| The Witcher 3 | 202010 - |
| Counter-Strike 2 | 352417 - |
| Far Cry 6 | 332616 - |
| Hogwarts Legacy | 171210 - |
| CoD: Modern Warfare III | 31271711 |
| Ghost of Tsushima | 191210 - |
| Cyberpunk 2077 | 17139 - |
| Shadow of the Tomb Raider | 292415 - |
| 1080p High | 27 |
| 1080p Ultra | 20 |
| 1440p Ultra | 13 |
| Margin of Error | Low |
| Time Spy | 2332 |
| Solar Bay | 9742 |
| Port Royal | 1010 |
| Fire Strike | 6909 |
| Wild Life Extreme | 4169 |
| Night Raid | 27654 |
| Background Blur | 73.8 img/sec |
| Face Detection | 30.1 img/sec |
| Horizon Detection | 0.93 Gpixels/sec |
| Edge Detection | 1.17 Gpixels/sec |
| Gaussian Blur | 1.17 Gpixels/sec |
| Feature Matching | 0.37 Gpixels/sec |
| Stereo Matching | 105 Gpixels/sec |
| Particle Physics | 4640 FPS |
| API | Vulkan |
| G2D Mark | 755 |
| DirectX 11 | 41 FPS |
| DirectX 12 | 20 FPS |
| GPU Compute | 2014 Ops/s |
| Aztec Ruins Normal Tier | 142 FPS |
| OS | Windows |
| Image Classification (SP) | 1976 |
| Image Segmentation (HP) | 3138 |
| Image Super Resolution (Q) | 6693 |
| Face Detection (HP) | 7213 |
| Pose Estimation (Q) | 21849 |
| Text Classification (SP) | 1497 |
| Machine Translation (HP) | 2579 |
| Object Detection (SP) | 2154 |
| Depth Estimation (Q) | 9635 |
| Style Transfer (SP) | 56269 |
| Framework | ONNX |
| Backend | DirectML |
| Case | Laptop |
| Purpose | Generic |
| Segment | Entry-level |
| Rival Equivalent | Radeon RX Vega 8 |
| Texture Mapping Units (TMUs) | 48 |
| Render Output Units (ROPs) | 32 |
| L1 Cache | 128KB per cluster |
| L2 Cache | 1024KB shared |
| Instructions Per Cycle | 2 IPC |
| Pixel Fill Rate | 70-77 GPixel/s |
| Texture Fill Rate | 106-115 GTexel/s |
| Transistor Density | 62.98 MTr/mm² |
| ECC | No |