Quick/Comparisons
The Radeon RX 7900 XTX outperforms the Intel Arc A770 in gaming due to its significantly higher score of 81, compared to the Intel Arc A770's 35.
The Intel Arc A770 takes the lead in content creation with a score of 43, narrowly surpassing the Radeon RX 7900 XTX's score of 75.
The scores for AI & Machine Learning are very close, with the Radeon RX 7900 XTX scoring 74 and the Intel Arc A770 scoring 31, resulting in a tie.
35 | Gaming | 81 |
43 | Workstation | 75 |
31 | AI/ML | 74 |
41 | Energy Efficiency | 43 |
37 | Quick Comparison Final Score | 77 |
| DG2-512 | Graphics Processor | Navi 31 |
| 4096 | Cores | 6144 |
| 256 / 128 | TMUs / ROPs | 384 / 192 |
| 2016.55 | Blender GPU | 4055.35 |
| 77 | Forza Horizon 5 | 174 |
| 124 | The Witcher 3 | 256 |
| 118 | Counter-Strike 2 | 232 |
| 97 | Far Cry 6 | 182 |
| 59 | Hogwarts Legacy | 137 |
| 79 | Call of Duty | 255 |
| 57 | Ghost of Tsushima | 138 |
| 80 | Cyberpunk 2077 | 181 |
| 106 | Tomb Raider | 277 |
| 89 | Average FPS | 204 |
| 106 | 1080p High | 241 |
| 89 | 1080p Ultra | 204 |
| 65 | 1440p Ultra | 159 |
| 37 | 4K Ultra | 94 |
| Low | Margin of Error | Low |
| 113935 | GB6 Compute Score | 224686 |
| 175 img/sec | Background Blur | 423.9 img/sec |
| 98 img/sec | Face Detection | 270.8 img/sec |
| 3.63 Gpixels/sec | Horizon Detection | 9.23 Gpixels/sec |
| 5.51 Gpixels/sec | Edge Detection | 13.7 Gpixels/sec |
| 8.51 Gpixels/sec | Gaussian Blur | 11.1 Gpixels/sec |
| 0.89 Gpixels/sec | Feature Matching | 1.74 Gpixels/sec |
| 492.6 Gpixels/sec | Stereo Matching | 871 Gpixels/sec |
| 24819.2 FPS | Particle Physics | 29891.2 FPS |
| OpenCL | API | OpenCL |
| 13358 | G3D Mark Score | 31406 |
| 761 | G2D Mark | 1285 |
| 78 FPS | DirectX 11 | 356 FPS |
| 75 FPS | DirectX 12 | 133 FPS |
| 5850 Ops/s | GPU Compute | 17768 Ops/s |
| 13344 | GB6 ML Single Precision | 42092 |
| 14119 | GB6 ML Half Precision | 50673 |
| 10082 | GB6 ML Quantized | 32874 |
| 5750 | Image Classification (SP) | 18608 |
| 16175 | Image Segmentation (HP) | 25385 |
| 12997 | Image Super Resolution (Q) | 45641 |
| 21409 | Face Detection (HP) | 71321 |
| 67820 | Pose Estimation (Q) | 257258 |
| 849 | Text Classification (SP) | 5240 |
| 2627 | Machine Translation (HP) | 7425 |
| 7465 | Object Detection (SP) | 21335 |
| 24186 | Depth Estimation (Q) | 66140 |
| 183238 | Style Transfer (SP) | 483867 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 307 GPixel/s | Pixel Fill Rate | 480 GPixel/s |
| 614 GTexel/s | Texture Fill Rate | 959 GTexel/s |
| 19.7 TFLOPS | FLOPS (FP32) | 61.4 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 16 GB | Memory Size | 24 GB |
| 2000 MHz | Memory Clock | 2500 MHz |
| 16000 Mbps | Effective Memory Speed | 20000 Mbps |
| 256-bit | Bus | 384-bit |
| No | ECC | No |
| 512 GB/s | Memory Bandwidth | 960 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 225 W | TGP | 355 W |
| TSMC | Manufacturing | TSMC |
| 6 nm | Fabrication Process | 5 nm |
| 406 mm² | Die Size | 529 mm² |
| 21 billion | Transistor Count | 57 billion |
| 51.72 MTr/mm² | Transistor Density | 107.75 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.2 |
| — | CUDA | — |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| — | DisplayPort | 2.1 |
| Intel | Vendor | Amd |
| Discrete | Build | Discrete |
| October 12, 2022 | Released | November 3, 2022 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Generation 12.7 | Architecture | RDNA 3.0 |
| DG2-512 | GPU Codename | Navi 31 |
| -Intel Core i5 12500or above | Recommended CPU | -Intel Core i7 13700Kor above |
| 12648 | Steel Nomad Lite Score | 29870 |
| 13554 | Time Spy | 30766 |
| 51114 | Solar Bay | 123758 |
| 7231 | Port Royal | 16749 |
| 34163 | Fire Strike | 73217 |
| 26266 | Wild Life Extreme | 58530 |
| 122588 | Night Raid | 184767 |
| 2100 MHz | Base Clock | 1929 MHz |
| 2400 MHz | Boost Clock | 2498 MHz |
| 4096 | Shading Units | 6144 |
| 256 | Texture Mapping Units (TMUs) | 384 |
| 128 | Render Output Units (ROPs) | 192 |
| — | Compute Units (Pipelines) | — |
| 512 | Tensor Cores | — |
| 32 | Ray-tracing Cores | 96 |
| — | L1 Cache | 256KB per cluster |
| 16MB shared | L2 Cache | 6MB shared |
| 2 IPC | Instructions Per Cycle | 4 IPC |
Compare with something else?
Compare Intel Arc A770 with another →