Radeon RX 9070
- Graphics Processor: Navi 48
- Cores: 3584
- TMUs / ROPs: 224 / 128
Review
Compare Radeon RX 9070 with other GPUs
Games
| Forza Horizon 5 | 229 156 137 102 |
| The Witcher 3 | 240 216 159 84 |
| Counter-Strike 2 | 307 215 158 84 |
| Far Cry 6 | 192 163 148 95 |
| Hogwarts Legacy | 146 119 97 52 |
| CoD: Modern Warfare III | 219 212 157 107 |
| Ghost of Tsushima | 146 122 105 65 |
| Cyberpunk 2077 | 174 149 102 47 |
| Shadow of the Tomb Raider | 262 248 188 99 |
| 1080p High | 213 |
| 1080p Ultra | 178 |
| 1440p Ultra | 139 |
| 4K Ultra | 82 |
| Margin of Error | Low |
Test in Benchmarks
Performance tests in graphics benchmarks and applications3D Mark
| Time Spy | 26693 |
| Solar Bay | 106118 |
| Port Royal | 15939 |
| Fire Strike | 60053 |
| Wild Life Extreme | 48323 |
| Night Raid | 177554 |
GeekBench 6 OpenCL
| Background Blur | 377.5 img/sec |
| Face Detection | 221.3 img/sec |
| Horizon Detection | 6.4 Gpixels/sec |
| Edge Detection | 8.99 Gpixels/sec |
| Gaussian Blur | 8.52 Gpixels/sec |
| Feature Matching | 1.36 Gpixels/sec |
| Stereo Matching | 606 Gpixels/sec |
| Particle Physics | 8521.2 FPS |
| API | OpenCL |
Passmark Graphics
| G2D Mark | 1262 |
| DirectX 11 | 280 FPS |
| DirectX 12 | 73 FPS |
| GPU Compute | 14799 Ops/s |
Artificial Intelligence Tests
Performance in machine learning and artificial intelligence tasksGeekBench 6 ML
| Image Classification (SP) | 15375 |
| Image Segmentation (HP) | 37834 |
| Image Super Resolution (Q) | 43910 |
| Face Detection (HP) | 67858 |
| Pose Estimation (Q) | 172939 |
| Text Classification (SP) | 4469 |
| Machine Translation (HP) | 8499 |
| Object Detection (SP) | 15393 |
| Depth Estimation (Q) | 66890 |
| Style Transfer (SP) | 365633 |
| Framework | ONNX |
| Backend | DirectML |
Specifications
Detailed technical specifications of the Radeon RX 9070General
| Vendor | Amd |
| Build | Discrete |
| Released | January 7, 2025 |
| Launch price (MSRP) | $549 |
| Case | Desktop |
| Purpose | Gaming |
| Segment | High-end |
| Architecture | RDNA 4.0 |
| GPU Codename | Navi 48 |
| Nvidia Equivalent | - GeForce RTX 5070 |
| Recommended CPU | - Intel Core Ultra 5 245K or above |
Desktop GPU rating - 16th place
Graphics Processing Unit
| Base Clock | 1330 MHz |
| Boost Clock | 2520 MHz |
| Shading Units | 3584 |
| Texture Mapping Units (TMUs) | 224 |
| Render Output Units (ROPs) | 128 |
| Compute Units (Pipelines) | 56 |
| Tensor Cores | 112 |
| Ray-tracing Cores | 56 |
| L1 Cache | 128KB per cluster |
| L2 Cache | 8MB shared |
| Instructions Per Cycle | 4 IPC |
Raw Performance
| Pixel Fill Rate | 323 GPixel/s |
| Texture Fill Rate | 564 GTexel/s |
Physical
| Interface | PCIe 5.0 x16 |
| TGP | 220 W |
| Manufacturing | TSMC |
| Fabrication Process | 4 nm |
| Die Size | 357 mm² |
| Transistor Count | 53.9 billion |
| Transistor Density | 150.98 MTr/mm² |
Memory
| Memory Type | GDDR6 |
| Memory Size | 16 GB |
| Memory Clock | 2518 MHz |
| Effective Memory Speed | 20100 Mbps |
| Bus | 256-bit |
| ECC | No |
API
| DirectX | 12.2 |
| Vulkan | 1.4 |
| OpenGL | 4.6 |
| OpenCL | 2.2 |
| CUDA | No |
| Ray Tracing | Yes |
| DLSS | No |
| DisplayPort | 2.1a |