Quick/Comparisons
The Radeon RX 6500M takes the lead in gaming due to its higher score of 14, thanks to its improved performance in this area.
The Radeon RX 6500M edges out the GeForce GTX 960 in content creation with a score of 21, showcasing its better performance in this area.
The Radeon RX 6500M outperforms the GeForce GTX 960 in AI/ML with a score of 14, thanks to its dedicated hardware and improved performance in this area.
8 | Gaming | 14 |
16 | Workstation | 21 |
6 | AI/ML |
Compare with something else?
Compare GeForce GTX 960 with another →14 |
18 | Energy Efficiency | 78 |
11 | Quick Comparison Final Score | 27 |
| GM206 | Graphics Processor | Navi 24 |
| 1024 | Cores | 1024 |
| 64 / 32 | TMUs / ROPs | 64 / 32 |
| 170.84 | Blender GPU | 384.34 |
| 14 | Forza Horizon 5 | 36 |
| 18 | The Witcher 3 | 38 |
| 29 | Counter-Strike 2 | 45 |
| 27 | Far Cry 6 | 44 |
| 12 | Hogwarts Legacy | 24 |
| 24 | Call of Duty | 39 |
| - | Ghost of Tsushima | 19 |
| 11 | Cyberpunk 2077 | 27 |
| 26 | Tomb Raider | 46 |
| 19 | Average FPS | 35 |
| 26 | 1080p High | 46 |
| 19 | 1080p Ultra | 35 |
| 13 | 1440p Ultra | 23 |
| - | 4K Ultra | 12 |
| Low | Margin of Error | Low |
| 20370 | GB6 Compute Score | 46706 |
| 51.7 img/sec | Background Blur | 112.4 img/sec |
| 26.1 img/sec | Face Detection | 61.7 img/sec |
| 0.84 Gpixels/sec | Horizon Detection | 1.49 Gpixels/sec |
| 1.36 Gpixels/sec | Edge Detection | 2.01 Gpixels/sec |
| 1.26 Gpixels/sec | Gaussian Blur | 1.81 Gpixels/sec |
| 0.14 Gpixels/sec | Feature Matching | 0.58 Gpixels/sec |
| 64.9 Gpixels/sec | Stereo Matching | 183 Gpixels/sec |
| 1825 FPS | Particle Physics | 6313.2 FPS |
| OpenCL | API | OpenCL |
| 6141 | G3D Mark Score | 7346 |
| 677 | G2D Mark | 371 |
| 42 FPS | DirectX 11 | 69 FPS |
| 28 FPS | DirectX 12 | 33 FPS |
| 2801 Ops/s | GPU Compute | 2545 Ops/s |
| 3784 | GB6 ML Single Precision | 8067 |
| 3830 | GB6 ML Half Precision | 9896 |
| 2724 | GB6 ML Quantized | 5967 |
| 1492 | Image Classification (SP) | 2972 |
| 2183 | Image Segmentation (HP) | 4132 |
| 3618 | Image Super Resolution (Q) | 9120 |
| 3654 | Face Detection (HP) | 11506 |
| 11306 | Pose Estimation (Q) | 35378 |
| 1315 | Text Classification (SP) | 1536 |
| 2085 | Machine Translation (HP) | 1869 |
| 1927 | Object Detection (SP) | 3259 |
| 5491 | Depth Estimation (Q) | 11922 |
| 22683 | Style Transfer (SP) | 81312 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 38 GPixel/s | Pixel Fill Rate | 70 GPixel/s |
| 75 GTexel/s | Texture Fill Rate | 140 GTexel/s |
| 2.4 TFLOPS | FLOPS (FP32) | 4.5 TFLOPS |
| GDDR5 | Memory Type | GDDR6 |
| 2 GB | Memory Size | 4 GB |
| 1753 MHz | Memory Clock | 2000 MHz |
| 7000 Mbps | Effective Memory Speed | 16000 Mbps |
| 128-bit | Bus | 64-bit |
| No | ECC | No |
| 112.2 GB/s | Memory Bandwidth | 128 GB/s |
| PCIe 3.0 x16 | Interface | PCIe 4.0 x4 |
| 120 W | TGP | 50 W |
| TSMC | Manufacturing | TSMC |
| 28 nm | Fabrication Process | 6 nm |
| 228 mm² | Die Size | 107 mm² |
| 2 billion | Transistor Count | 5.4 billion |
| 8.77 MTr/mm² | Transistor Density | 50.47 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.2 |
| 5.2 | CUDA | — |
| No | Ray Tracing | Yes |
| No | DLSS | No |
| 1.2 | DisplayPort | 1.4 |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| January 22, 2015 | Released | January 4, 2022 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | Entry-level |
| Maxwell 2.0 | Architecture | RDNA 2.0 |
| GM206 | GPU Codename | Navi 24 |
| — | Recommended CPU | -AMD Ryzen 5 6600Hor above |
| 2179 | Steel Nomad Lite Score | 3850 |
| 2302 | Time Spy | 3831 |
| — | Solar Bay | 15034 |
| — | Port Royal | 248 |
| 7720 | Fire Strike | 12343 |
| 4543 | Wild Life Extreme | 8150 |
| 31114 | Night Raid | 45550 |
| 1127 MHz | Base Clock | 2000 MHz |
| 1178 MHz | Boost Clock | 2190 MHz |
| 1024 | Shading Units | 1024 |
| 64 | Texture Mapping Units (TMUs) | 64 |
| 32 | Render Output Units (ROPs) | 32 |
| 8 | Compute Units (Pipelines) | 16 |
| No | Tensor Cores | — |
| No | Ray-tracing Cores | 16 |
| 48KB per cluster | L1 Cache | 128KB per cluster |
| 1MB shared | L2 Cache | 1MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |