Quick/Comparisons
The GeForce RTX 5090's 100 score in gaming performance far surpasses the Radeon RX 7600M XT's 30, making it the clear winner for gaming enthusiasts.
The GeForce RTX 5090's workstation performance score of 93 is significantly higher than the Radeon RX 7600M XT's 33, giving it the edge in content creation.
The GeForce RTX 5090's AI/ML compute capabilities, with a score of 95, leave the Radeon RX 7600M XT's 31 score in the dust, making it the winner for AI and machine learning tasks.
100 | Gaming | 30 |
93 | Workstation | 33 |
95 | AI/ML | 31 |
49 | Energy Efficiency | 59 |
96 | Quick Comparison Final Score | 36 |
| GB202-300 | Graphics Processor | Navi 33 |
| 21760 | Cores | 2048 |
| 680 / 176 | TMUs / ROPs | 128 / 64 |
| 15010.34 | Blender GPU | 1053.82 |
| 244 | Forza Horizon 5 | 67 |
| 323 | The Witcher 3 | 91 |
| 265 | Counter-Strike 2 | 126 |
| 202 | Far Cry 6 | 85 |
| 207 | Hogwarts Legacy | 56 |
| 327 | Call of Duty | 80 |
| 179 | Ghost of Tsushima | 58 |
| 229 | Cyberpunk 2077 | 61 |
| 299 | Tomb Raider | 102 |
| 253 | Average FPS | 81 |
| 294 | 1080p High | 102 |
| 253 | 1080p Ultra | 81 |
| 207 | 1440p Ultra | 57 |
| 124 | 4K Ultra | 30 |
| Low | Margin of Error | Low |
| 419309 | GB6 Compute Score | 77586 |
| 333 img/sec | Background Blur | 186.5 img/sec |
| 298.5 img/sec | Face Detection | 104.8 img/sec |
| 21.7 Gpixels/sec | Horizon Detection | 2.79 Gpixels/sec |
| 33.7 Gpixels/sec | Edge Detection | 3.91 Gpixels/sec |
| 34.5 Gpixels/sec | Gaussian Blur | 3.3 Gpixels/sec |
| 2.53 Gpixels/sec | Feature Matching | 0.87 Gpixels/sec |
| 2850 Gpixels/sec | Stereo Matching | 253.1 Gpixels/sec |
| 59146.9 FPS | Particle Physics | 9459.7 FPS |
| OpenCL | API | OpenCL |
| 38955 | G3D Mark Score | 14720 |
| 1413 | G2D Mark | 938 |
| 341 FPS | DirectX 11 | 138 FPS |
| 178 FPS | DirectX 12 | 63 FPS |
| 24632 Ops/s | GPU Compute | 7340 Ops/s |
| 114685 | Face Detection (HP) | 26407 |
| 344729 | Pose Estimation (Q) | 87536 |
| 4464 | Text Classification (SP) | 3561 |
| 8184 | Machine Translation (HP) | 2851 |
| 27802 | Object Detection (SP) | 7042 |
| 85572 | Depth Estimation (Q) | 40441 |
| 687080 | Style Transfer (SP) | 175589 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 53861 | GB6 ML Single Precision | 17923 |
| 81080 | GB6 ML Half Precision | 21900 |
| 38938 | GB6 ML Quantized | 14986 |
| 18948 | Image Classification (SP) | 8974 |
| 55109 | Image Segmentation (HP) | 8862 |
| 58478 | Image Super Resolution (Q) | 23799 |
| 424 GPixel/s | Pixel Fill Rate | 158 GPixel/s |
| 1637 GTexel/s | Texture Fill Rate | 316 GTexel/s |
| 104.8 TFLOPS | FLOPS (FP32) | 20.2 TFLOPS |
| GDDR7 | Memory Type | GDDR6 |
| 32 GB | Memory Size | 8 GB |
| 1750 MHz | Memory Clock | 2250 MHz |
| 28000 Mbps | Effective Memory Speed | 18000 Mbps |
| 512-bit | Bus | 128-bit |
| No | ECC | No |
| 1792 GB/s | Memory Bandwidth | 288 GB/s |
| PCIe 5.0 x16 | Interface | PCIe 4.0 x16 |
| 575 W | TGP | 120 W |
| TSMC | Manufacturing | TSMC |
| 4 nm | Fabrication Process | 6 nm |
| 750 mm² | Die Size | 204 mm² |
| 92.2 billion | Transistor Count | 13 billion |
| 122.93 MTr/mm² | Transistor Density | 63.73 MTr/mm² |
| 12.2 | DirectX | 12 |
| 1.4 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.2 |
| 12 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 4 | DLSS | No |
| 2.1b | DisplayPort | — |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| January 7, 2025 | Released | January 4, 2023 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Gaming |
| High-end | Segment | Mid-range |
| Blackwell 2.0 | Architecture | RDNA 3.0 |
| GB202-300 | GPU Codename | Navi 33 |
| -Intel Core Ultra 9 285Kor above | Recommended CPU | -Intel Core i5 12400or above |
| 52638 | Steel Nomad Lite Score | 8599 |
| 46941 | Time Spy | 8788 |
| 235224 | Solar Bay | 35459 |
| 38113 | Port Royal | 4658 |
| 87433 | Fire Strike | 23974 |
| 108736 | Wild Life Extreme | 15920 |
| 207058 | Night Raid | 61463 |
| 2017 MHz | Base Clock | 1280 MHz |
| 2407 MHz | Boost Clock | 2469 MHz |
| 21760 | Shading Units | 2048 |
| 680 | Texture Mapping Units (TMUs) | 128 |
| 176 | Render Output Units (ROPs) | 64 |
| 170 | Compute Units (Pipelines) | — |
| 680 | Tensor Cores | — |
| 170 | Ray-tracing Cores | 32 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 96MB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 4 IPC |
Compare with something else?
Compare GeForce RTX 5090 with another →