Quick/Comparisons
The Radeon RX 7900 XTX takes the lead in gaming performance with an 81 score, thanks to its exceptional frame rates and smooth visuals.
The Radeon RX 7900 XTX edges out the Radeon RX 7900M in content creation with a 75 score, driven by its capable workstation performance and large memory pool.
The Radeon RX 7900 XTX outperforms the Radeon RX 7900M in AI/ML compute with a 74 score, leveraging its RDNA 3.0 architecture and 24GB memory for efficient processing.
81 | Gaming | 54 |
75 | Workstation | 55 |
74 | AI/ML | 45 |
43 | Energy Efficiency | 45 |
77 | Quick Comparison Final Score | 52 |
| Navi 31 | Graphics Processor | Navi 31 |
| 6144 | Cores | 4608 |
| 384 / 192 | TMUs / ROPs | 288 / 192 |
| 4055.35 | Blender GPU | 2436.73 |
| 174 | Forza Horizon 5 | 129 |
| 256 | The Witcher 3 | 164 |
| 232 | Counter-Strike 2 | 197 |
| 182 | Far Cry 6 | 140 |
| 137 | Hogwarts Legacy | 96 |
| 255 | Call of Duty | 151 |
| 138 | Ghost of Tsushima | 96 |
| 181 | Cyberpunk 2077 | 109 |
| 277 | Tomb Raider | 182 |
| 204 | Average FPS | 140 |
| 241 | 1080p High | 171 |
| 204 | 1080p Ultra | 140 |
| 159 | 1440p Ultra | 107 |
| 94 | 4K Ultra | 63 |
| Low | Margin of Error | High |
| 224686 | GB6 Compute Score | 151924 |
| 423.9 img/sec | Background Blur | 305.8 img/sec |
| 270.8 img/sec | Face Detection | 201.4 img/sec |
| 9.23 Gpixels/sec | Horizon Detection | 5.87 Gpixels/sec |
| 13.7 Gpixels/sec | Edge Detection | 8.19 Gpixels/sec |
| 11.1 Gpixels/sec | Gaussian Blur | 6.77 Gpixels/sec |
| 1.74 Gpixels/sec | Feature Matching | 1.32 Gpixels/sec |
| 871 Gpixels/sec | Stereo Matching | 540.4 Gpixels/sec |
| 29891.2 FPS | Particle Physics | 22269 FPS |
| OpenCL | API | OpenCL |
| 31406 | G3D Mark Score | 22111 |
| 1285 | G2D Mark | 893 |
| 356 FPS | DirectX 11 | 256 FPS |
| 133 FPS | DirectX 12 | 96 FPS |
| 17768 Ops/s | GPU Compute | 9043 Ops/s |
| 42092 | GB6 ML Single Precision | 25398 |
| 50673 | GB6 ML Half Precision | 28752 |
| 32874 | GB6 ML Quantized | 19862 |
| 18608 | Image Classification (SP) | 11249 |
| 25385 | Image Segmentation (HP) | 13771 |
| 45641 | Image Super Resolution (Q) | 30143 |
| 71321 | Face Detection (HP) | 31918 |
| 257258 | Pose Estimation (Q) | 163344 |
| 5240 | Text Classification (SP) | 3882 |
| 7425 | Machine Translation (HP) | 3734 |
| 21335 | Object Detection (SP) | 12629 |
| 66140 | Depth Estimation (Q) | 45969 |
| 483867 | Style Transfer (SP) | 303544 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 480 GPixel/s | Pixel Fill Rate | 401 GPixel/s |
| 959 GTexel/s | Texture Fill Rate | 602 GTexel/s |
| 61.4 TFLOPS | FLOPS (FP32) | 38.5 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 24 GB | Memory Size | 16 GB |
| 2500 MHz | Memory Clock | 2250 MHz |
| 20000 Mbps | Effective Memory Speed | 18000 Mbps |
| 384-bit | Bus | 256-bit |
| No | ECC | No |
| 960 GB/s | Memory Bandwidth | 576 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 355 W | TGP | 180 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 5 nm |
| 529 mm² | Die Size | 529 mm² |
| 57 billion | Transistor Count | 57.7 billion |
| 107.75 MTr/mm² | Transistor Density | 109.07 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 2.2 | OpenCL | 2.2 |
| — | CUDA | — |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| 2.1 | DisplayPort | — |
| Amd | Vendor | Amd |
| Discrete | Build | Discrete |
| November 3, 2022 | Released | October 19, 2023 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Gaming |
| High-end | Segment | High-end |
| RDNA 3.0 | Architecture | RDNA 3.0 |
| Navi 31 | GPU Codename | Navi 31 |
| -Intel Core i7 13700Kor above | Recommended CPU | -Intel Core i7 13700HXor above |
| 29870 | Steel Nomad Lite Score | 17147 |
| 30766 | Time Spy | 18071 |
| 123758 | Solar Bay | 77397 |
| 16749 | Port Royal | 10145 |
| 73217 | Fire Strike | 44044 |
| 58530 | Wild Life Extreme | 35258 |
| 184767 | Night Raid | 153739 |
| 1929 MHz | Base Clock | 1825 MHz |
| 2498 MHz | Boost Clock | 2090 MHz |
| 6144 | Shading Units | 4608 |
| 384 | Texture Mapping Units (TMUs) | 288 |
| 192 | Render Output Units (ROPs) | 192 |
| — | Compute Units (Pipelines) | — |
| — | Tensor Cores | — |
| 96 | Ray-tracing Cores | 72 |
| 256KB per cluster | L1 Cache | 256KB per cluster |
| 6MB shared | L2 Cache | 6MB shared |
| 4 IPC | Instructions Per Cycle | 4 IPC |
Compare with something else?
Compare Radeon RX 7900 with another →