Key Differences
Advantages of Intel Arc B580
Advantages of Radeon RX 9070
Specifications
Performance
| TGP | 190 W | 220 W |
Storage Details
| Bus | 192-bit | 256-bit |
Ports
| DisplayPort | 2.1 | 2.1a |
Other
| Forza Horizon 5 | 123786751 | 227155137103 |
| The Witcher 3 | 1471289146 | 24721315682 |
| Counter-Strike 2 | 1621168340 | 30921716181 |
| Far Cry 6 | 108987643 | 19016914894 |
| Hogwarts Legacy | 71574426 | 1461199752 |
| CoD: Modern Warfare III | 82856138 | 216213158106 |
| Ghost of Tsushima | 72614225 | 15112610568 |
| Cyberpunk 2077 | 90825526 | 16815410350 |
| Shadow of the Tomb Raider | 1241078343 | 26424718699 |
| 1080p High | 109 | 213 |
| 1080p Ultra | 90 | 179 |
| 1440p Ultra | 67 | 139 |
| 4K Ultra | 38 | 82 |
| Margin of Error | Low | Low |
| Time Spy | 14654 | 26999 |
| Solar Bay | 55346 | 105011 |
| Port Royal | 7910 | 16147 |
| Fire Strike | 36715 | 60580 |
| Wild Life Extreme | 28283 | 48008 |
| Night Raid | 123579 | 173193 |
| Background Blur | 284.6 img/sec | 377.5 img/sec |
| Face Detection | 150 img/sec | 221.3 img/sec |
| Horizon Detection | 2.67 Gpixels/sec | 6.4 Gpixels/sec |
| Edge Detection | 5.22 Gpixels/sec | 8.99 Gpixels/sec |
| Gaussian Blur | 6.35 Gpixels/sec | 8.52 Gpixels/sec |
| Feature Matching | 1.18 Gpixels/sec | 1.36 Gpixels/sec |
| Stereo Matching | 397 Gpixels/sec | 606 Gpixels/sec |
| Particle Physics | 15177.3 FPS | 8521.2 FPS |
| API | OpenGL | OpenGL |
| G2D Mark | 703 | 1277 |
| DirectX 11 | 129 FPS | 279 FPS |
| DirectX 12 | 77 FPS | 74 FPS |
| GPU Compute | 7865 Ops/s | 14845 Ops/s |
| Aztec Ruins Normal Tier | 847.7 FPS | 1412.6 FPS |
| Car Chase | 779.3 FPS | 1332.4 FPS |
| Manhattan 3.1.1 (1440p) | 608.2 FPS | 1061.2 FPS |
| T-Rex | 2266.1 FPS | 4127 FPS |
| Tessellation | 4148.9 FPS | 6097.5 FPS |
| ALU 2 | 3674.6 FPS | 7774.7 FPS |
| Driver Overhead 2 | 184.2 FPS | 276.1 FPS |
| OS | Windows | Windows |
| Image Classification (SP) | 10654 | 15375 |
| Image Segmentation (HP) | 18608 | 37834 |
| Image Super Resolution (Q) | 16715 | 43910 |
| Face Detection (HP) | 40348 | 67858 |
| Pose Estimation (Q) | 64494 | 172939 |
| Text Classification (SP) | 3091 | 4469 |
| Machine Translation (HP) | 3530 | 8499 |
| Object Detection (SP) | 13690 | 15393 |
| Depth Estimation (Q) | 36797 | 66890 |
| Style Transfer (SP) | 197057 | 365633 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
| Vendor | Intel | Amd |
| Build | Discrete | Discrete |
| Released | December 3, 2024 | January 7, 2025 |
| Launch price (MSRP) | $249 | $549 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | Mid-range | High-end |
| Architecture | Generation 12.7 | RDNA 4.0 |
| GPU Codename | BMG-G21 | Navi 48 |
| Rival Equivalent | GeForce RTX 4060 | GeForce RTX 5070 |
| Base Clock | 1700 MHz | 1330 MHz |
| Boost Clock | 2670 MHz | 2520 MHz |
| Shading Units | 2560 | 3584 |
| Texture Mapping Units (TMUs) | 160 | 224 |
| Render Output Units (ROPs) | 80 | 128 |
| Compute Units (Pipelines) | 20 | 56 |
| Tensor Cores | 160 | 112 |
| Ray-tracing Cores | 20 | 56 |
| L1 Cache | 256KB per cluster | 128KB per cluster |
| L2 Cache | 12MB shared | 4MB shared |
| Instructions Per Cycle | 2 IPC | 4 IPC |
| Pixel Fill Rate | 214 GPixel/s | 323 GPixel/s |
| Texture Fill Rate | 427 GTexel/s | 564 GTexel/s |
| Interface | PCIe 4.0 x8 | PCIe 5.0 x16 |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 5 nm | 4 nm |
| Die Size | 272 mm² | 356.5 mm² |
| Transistor Count | 19.6 billion | 53.9 billion |
| Transistor Density | 72.06 MTr/mm² | 151.19 MTr/mm² |
| Max. Temperature | - | - |
| Memory Type | GDDR6 | GDDR6 |
| Memory Size | 12 GB | 16 GB |
| Memory Clock | 2375 MHz | 2518 MHz |
| Effective Memory Speed | 19000 Mbps | 20100 Mbps |
| ECC | No | No |
| DirectX | 12 | 12 |
| Vulkan | 1.3 | 1.3 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 2.2 |
| CUDA | No | No |
| Ray Tracing | Yes | Yes |
| DLSS | No | No |