Quick/Comparisons
The Radeon RX 7900 XT's superior gaming performance, with a 25-point lead, makes it the clear winner in this category. Its 72 score far surpasses the Nvidia RTX 4000 Ada's 47.
Although the Radeon RX 7900 XT's 18-point lead is significant, the Nvidia RTX 4000 Ada's 52 score is still respectable, but its lower score means it falls short in this category. The Nvidia RTX 4000 Ada's 52 score is not enough to surpass the Radeon RX 7900 XT's 70.
The Radeon RX 7900 XT's 24-point lead in AI/ML performance makes it the clear winner in this category. Its 68 score far surpasses the Nvidia RTX 4000 Ada's 44.
47 | Gaming | 72 |
52 | Workstation | 70 |
44 | AI/ML | 68 |
59 | Energy Efficiency | 45 |
48 | Quick Comparison Final Score | 70 |
| AD104 | Graphics Processor | Navi 31 |
| 6144 | Cores | 5376 |
| 192 / 80 | TMUs / ROPs | 336 / 192 |
| 4481.44 | Blender GPU | 3548.59 |
| 98 | Forza Horizon 5 | 157 |
| 143 | The Witcher 3 | 228 |
| 201 | Counter-Strike 2 | 223 |
| 123 | Far Cry 6 | 173 |
| 82 | Hogwarts Legacy | 126 |
| 119 | Call of Duty | 222 |
| 79 | Ghost of Tsushima | 129 |
| 96 | Cyberpunk 2077 | 158 |
| 157 | Tomb Raider | 247 |
| 122 | Average FPS | 185 |
| 147 | 1080p High | 219 |
| 122 | 1080p Ultra | 185 |
| 92 | 1440p Ultra | 144 |
| 53 | 4K Ultra | 85 |
| High | Margin of Error | Low |
| 151680 | GB6 Compute Score | 203754 |
| 235.3 img/sec | Background Blur | 419.8 img/sec |
| 157.9 img/sec | Face Detection | 240.5 img/sec |
| 5.53 Gpixels/sec | Horizon Detection | 8.72 Gpixels/sec |
| 7.44 Gpixels/sec | Edge Detection | 12 Gpixels/sec |
| 7.87 Gpixels/sec | Gaussian Blur | 9.88 Gpixels/sec |
| 1.58 Gpixels/sec | Feature Matching | 1.71 Gpixels/sec |
| 737.7 Gpixels/sec | Stereo Matching | 701.7 Gpixels/sec |
| 22409.2 FPS | Particle Physics | 26631.4 FPS |
| OpenCL | API | OpenCL |
| 23716 | G3D Mark Score | 29015 |
| 1118 | G2D Mark | 1241 |
| 204 FPS | DirectX 11 | 340 FPS |
| 93 FPS | DirectX 12 | 111 FPS |
| 11936 Ops/s | GPU Compute | 16851 Ops/s |
| 22372 | GB6 ML Single Precision | 39099 |
| 35651 | GB6 ML Half Precision | 47298 |
| 17081 | GB6 ML Quantized | 30875 |
| 9717 | Image Classification (SP) | 16951 |
| 19365 | Image Segmentation (HP) | 23234 |
| 29227 | Image Super Resolution (Q) | 41128 |
| 40416 | Face Detection (HP) | 67615 |
| 105639 | Pose Estimation (Q) | 237026 |
| 3577 | Text Classification (SP) | 4961 |
| 4193 | Machine Translation (HP) | 7043 |
| 10820 | Object Detection (SP) | 18737 |
| 36574 | Depth Estimation (Q) | 67746 |
| 200936 | Style Transfer (SP) | 457919 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 174 GPixel/s | Pixel Fill Rate | 460 GPixel/s |
| 418 GTexel/s | Texture Fill Rate | 804 GTexel/s |
| 26.7 TFLOPS | FLOPS (FP32) | 51.5 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 20 GB | Memory Size | 20 GB |
| 2250 MHz | Memory Clock | 2500 MHz |
| 18000 Mbps | Effective Memory Speed | 20000 Mbps |
| 160-bit | Bus | 320-bit |
| No | ECC | No |
| 360 GB/s | Memory Bandwidth | 800 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 130 W | TGP | 300 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 5 nm |
| 294 mm² | Die Size | 529 mm² |
| 35 billion | Transistor Count | 57 billion |
| 119.05 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 |
| 8.9 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | No |
| 1.4a | DisplayPort | 2.1 |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| August 9, 2023 | Released | December 13, 2023 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | High-end |
| Ada Lovelace | Architecture | RDNA 3.0 |
| AD104 | GPU Codename | Navi 31 |
| -Intel Core i7 14700Kor above | Recommended CPU | -Intel Core i7 13700Kor above |
| 14674 | Steel Nomad Lite Score | 24675 |
| 14438 | Time Spy | 27067 |
| 65875 | Solar Bay | 104912 |
| 8955 | Port Royal | 14358 |
| 37363 | Fire Strike | 67256 |
| — | Wild Life Extreme | 48319 |
| 158480 | Night Raid | 176267 |
| 1500 MHz | Base Clock | 1387 MHz |
| 2175 MHz | Boost Clock | 2394 MHz |
| 6144 | Shading Units | 5376 |
| 192 | Texture Mapping Units (TMUs) | 336 |
| 80 | Render Output Units (ROPs) | 192 |
| 48 | Compute Units (Pipelines) | — |
| 192 | Tensor Cores | — |
| 48 | Ray-tracing Cores | 84 |
| 128KB per cluster | L1 Cache | 256KB per cluster |
| 48MB shared | L2 Cache | 6MB shared |
| 2 IPC | Instructions Per Cycle | 4 IPC |
Compare with something else?
Compare Nvidia RTX 4000 with another →