Quick/Comparisons
11 | Gaming | 64 |
20 | Workstation | 60 |
13 | AI/ML |
Compare with something else?
Compare GeForce GTX 1650 with another →59 |
58 | Energy Efficiency | 44 |
15 | Quick Comparison Final Score | 62 |
| TU117 | Graphics Processor | Navi 31 |
| 896 | Cores | 5120 |
| 56 / 32 | TMUs / ROPs | 320 / 160 |
| 443.47 | Blender GPU | 2808.96 |
| 32 | Forza Horizon 5 | 143 |
| 33 | The Witcher 3 | 201 |
| 54 | Counter-Strike 2 | 217 |
| 41 | Far Cry 6 | 155 |
| 21 | Hogwarts Legacy | 109 |
| 31 | Call of Duty | 190 |
| 15 | Ghost of Tsushima | 116 |
| 22 | Cyberpunk 2077 | 138 |
| 43 | Tomb Raider | 221 |
| 32 | Average FPS | 166 |
| 42 | 1080p High | 198 |
| 32 | 1080p Ultra | 166 |
| 22 | 1440p Ultra | 128 |
| 11 | 4K Ultra | 74 |
| Low | Margin of Error | Low |
| 39481 | GB6 Compute Score | 172224 |
| 66.4 img/sec | Background Blur | 352.4 img/sec |
| 43.3 img/sec | Face Detection | 229 img/sec |
| 1.51 Gpixels/sec | Horizon Detection | 6.42 Gpixels/sec |
| 1.68 Gpixels/sec | Edge Detection | 8.86 Gpixels/sec |
| 1.87 Gpixels/sec | Gaussian Blur | 7.89 Gpixels/sec |
| 0.49 Gpixels/sec | Feature Matching | 1.5 Gpixels/sec |
| 187.9 Gpixels/sec | Stereo Matching | 662.8 Gpixels/sec |
| 5261.5 FPS | Particle Physics | 24012 FPS |
| OpenCL | API | OpenCL |
| 7871 | G3D Mark Score | 27367 |
| 553 | G2D Mark | 1207 |
| 58 FPS | DirectX 11 | 300 FPS |
| 35 FPS | DirectX 12 | 110 FPS |
| 3036 Ops/s | GPU Compute | 15171 Ops/s |
| 7569 | GB6 ML Single Precision | 34533 |
| 5940 | GB6 ML Half Precision | 42224 |
| 5581 | GB6 ML Quantized | 26568 |
| 3208 | Image Classification (SP) | 15095 |
| 4254 | Image Segmentation (HP) | 21206 |
| 7050 | Image Super Resolution (Q) | 33962 |
| 6849 | Face Detection (HP) | 57614 |
| 23562 | Pose Estimation (Q) | 214748 |
| 1755 | Text Classification (SP) | 4325 |
| 3096 | Machine Translation (HP) | 6598 |
| 3710 | Object Detection (SP) | 16608 |
| 12946 | Depth Estimation (Q) | 57731 |
| 48894 | Style Transfer (SP) | 394987 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 53 GPixel/s | Pixel Fill Rate | 359 GPixel/s |
| 93 GTexel/s | Texture Fill Rate | 718 GTexel/s |
| 3 TFLOPS | FLOPS (FP32) | 46 TFLOPS |
| GDDR5 | Memory Type | GDDR6 |
| 4 GB | Memory Size | 16 GB |
| 2001 MHz | Memory Clock | 2250 MHz |
| 8000 Mbps | Effective Memory Speed | 18000 Mbps |
| 128-bit | Bus | 256-bit |
| No | ECC | No |
| 128.1 GB/s | Memory Bandwidth | 576 GB/s |
| PCIe 3.0 x16 | Interface | PCIe 4.0 x16 |
| 75 W | TGP | 260 W |
| TSMC | Manufacturing | TSMC |
| 12 nm | Fabrication Process | 5 nm |
| 200 mm² | Die Size | 529 mm² |
| 4 billion | Transistor Count | 57 billion |
| 20 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 |
| 7.5 | CUDA | — |
| No | Ray Tracing | Yes |
| No | DLSS | No |
| 1.4a | DisplayPort | 2.1 |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| April 23, 2019 | Released | — |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Entry-level | Segment | High-end |
| Turing | Architecture | RDNA 3.0 |
| TU117 | GPU Codename | Navi 31 |
| -Intel Core i5 9400or above | Recommended CPU | — |
| 3419 | Steel Nomad Lite Score | 21359 |
| 3563 | Time Spy | 22537 |
| 12275 | Solar Bay | 90757 |
| — | Port Royal | 12461 |
| 9696 | Fire Strike | 56212 |
| 7350 | Wild Life Extreme | 41740 |
| 46189 | Night Raid | 169038 |
| 1485 MHz | Base Clock | 1287 MHz |
| 1665 MHz | Boost Clock | 2245 MHz |
| 896 | Shading Units | 5120 |
| 56 | Texture Mapping Units (TMUs) | 320 |
| 32 | Render Output Units (ROPs) | 160 |
| 14 | Compute Units (Pipelines) | — |
| No | Tensor Cores | — |
| No | Ray-tracing Cores | 80 |
| 64KB per cluster | L1 Cache | 256KB per cluster |
| 1024KB shared | L2 Cache | 6MB shared |
| 2 IPC | Instructions Per Cycle | 4 IPC |