Quick/Comparisons
The Nvidia RTX 5000 Ada takes the lead in gaming performance, thanks to its 73 score, which is 7 points higher than the Radeon RX 9070's 66.
The Nvidia RTX 5000 Ada excels in workstation tasks, with an 82 score, significantly outpacing the Radeon RX 9070's 56.
The Radeon RX 9070 takes the lead in AI/ML tasks, with a 63 score, 9 points higher than the Nvidia RTX 5000 Ada's 54.
73 | Gaming | 66 |
82 | Workstation | 56 |
54 | AI/ML | 63 |
45 | Energy Efficiency | 49 |
73 | Quick Comparison Final Score | 62 |
| AD102 | Graphics Processor | Navi 48 |
| 12800 | Cores | 3584 |
| 400 / 176 | TMUs / ROPs | 224 / 128 |
| 7348.98 | Blender GPU | 2749.68 |
| 146 | Forza Horizon 5 | 156 |
| 186 | The Witcher 3 | 217 |
| 213 | Counter-Strike 2 | 218 |
| 151 | Far Cry 6 | 164 |
| 117 | Hogwarts Legacy | 123 |
| 179 | Call of Duty | 206 |
| 108 | Ghost of Tsushima | 123 |
| 148 | Cyberpunk 2077 | 149 |
| 242 | Tomb Raider | 240 |
| 166 | Average FPS | 177 |
| 196 | 1080p High | 213 |
| 166 | 1080p Ultra | 177 |
| 132 | 1440p Ultra | 139 |
| 78 | 4K Ultra | 82 |
| High | Margin of Error | Low |
| 260099 | GB6 Compute Score | 149949 |
| 286.6 img/sec | Background Blur | 377.5 img/sec |
| 220.4 img/sec | Face Detection | 221.3 img/sec |
| 9.93 Gpixels/sec | Horizon Detection | 6.4 Gpixels/sec |
| 13.7 Gpixels/sec | Edge Detection | 8.99 Gpixels/sec |
| 18.9 Gpixels/sec | Gaussian Blur | 8.52 Gpixels/sec |
| 2.2 Gpixels/sec | Feature Matching | 1.36 Gpixels/sec |
| 1580 Gpixels/sec | Stereo Matching | 606 Gpixels/sec |
| 41235.6 FPS | Particle Physics | 8521.2 FPS |
| OpenCL | API | OpenCL |
| 30196 | G3D Mark Score | 25389 |
| 998 | G2D Mark | 1265 |
| 273 FPS | DirectX 11 | 280 FPS |
| 106 FPS | DirectX 12 | 74 FPS |
| 20736 Ops/s | GPU Compute | 14791 Ops/s |
| 24715 | GB6 ML Single Precision | 37056 |
| 36753 | GB6 ML Half Precision | 54866 |
| 19649 | GB6 ML Quantized | 29946 |
| 9943 | Image Classification (SP) | 15375 |
| 19696 | Image Segmentation (HP) | 37834 |
| 32252 | Image Super Resolution (Q) | 43910 |
| 43098 | Face Detection (HP) | 67858 |
| 138689 | Pose Estimation (Q) | 172939 |
| 2893 | Text Classification (SP) | 4469 |
| 4064 | Machine Translation (HP) | 8499 |
| 12467 | Object Detection (SP) | 15393 |
| 50352 | Depth Estimation (Q) | 66890 |
| 265482 | Style Transfer (SP) | 365633 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 449 GPixel/s | Pixel Fill Rate | 323 GPixel/s |
| 1020 GTexel/s | Texture Fill Rate | 564 GTexel/s |
| 65.3 TFLOPS | FLOPS (FP32) | 36.1 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 32 GB | Memory Size | 16 GB |
| 2250 MHz | Memory Clock | 2518 MHz |
| 18000 Mbps | Effective Memory Speed | 20100 Mbps |
| 256-bit | Bus | 256-bit |
| No | ECC | No |
| 576 GB/s | Memory Bandwidth | 640 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 250 W | TGP | 220 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 4 nm |
| 609 mm² | Die Size | 356.5 mm² |
| 76 billion | Transistor Count | 53.9 billion |
| 124.79 MTr/mm² | Transistor Density | 151.19 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.2 |
| 8.9 | CUDA | No |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | No |
| 1.4a | DisplayPort | 2.1a |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| August 9, 2023 | Released | January 7, 2025 |
| Desktop | Case | Desktop |
| Professional | Purpose | Gaming |
| High-end | Segment | High-end |
| Ada Lovelace | Architecture | RDNA 4.0 |
| AD102 | GPU Codename | Navi 48 |
| -Intel Core i9 14900Kor above | Recommended CPU | -Intel Core Ultra 5 245Kor above |
| 29190 | Steel Nomad Lite Score | 22783 |
| 18675 | Time Spy | 26826 |
| 128428 | Solar Bay | 105512 |
| 17754 | Port Royal | 16089 |
| 62027 | Fire Strike | 60226 |
| 52866 | Wild Life Extreme | 48184 |
| 210443 | Night Raid | 176216 |
| 1155 MHz | Base Clock | 1330 MHz |
| 2550 MHz | Boost Clock | 2520 MHz |
| 12800 | Shading Units | 3584 |
| 400 | Texture Mapping Units (TMUs) | 224 |
| 176 | Render Output Units (ROPs) | 128 |
| 100 | Compute Units (Pipelines) | 56 |
| 400 | Tensor Cores | 112 |
| 100 | Ray-tracing Cores | 56 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 72MB shared | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 4 IPC |
Compare with something else?
Compare Nvidia RTX 5000 with another →