Quick/Comparisons
A budget Nvidia GeForce GTX 1660 Ti graphics card with limited performance overall. Suited for casual gaming and basic graphics tasks.
| Gaming | 20 /100 |
| Workstation | 29 /100 |
| AI/ML | 18 /100 |
| Energy Efficiency | 44 |
Looking for more comparisons?
Compare GeForce GTX →| Quick Comparison Final Score | 23 /100 |
| Graphics Processor | TU116 |
| Cores | 1536 |
| TMUs / ROPs | 96 / 48 |
| Blender GPU | 796.71 |
| Forza Horizon 5 | 52 |
| The Witcher 3 | 69 |
| Counter-Strike 2 | 96 |
| Far Cry 6 | 64 |
| Hogwarts Legacy | 37 |
| Call of Duty | 47 |
| Ghost of Tsushima | 32 |
| Cyberpunk 2077 | 42 |
| Tomb Raider | 74 |
| Average FPS | 57 |
| 1080p High | 74 |
| 1080p Ultra | 57 |
| 1440p Ultra | 41 |
| 4K Ultra | 21 |
| Margin of Error | Low |
| GB6 Compute Score | 72340 |
| Background Blur | 124.8 img/sec |
| Face Detection | 69.4 img/sec |
| Horizon Detection | 2.67 Gpixels/sec |
| Edge Detection | 3.48 Gpixels/sec |
| Gaussian Blur | 3.52 Gpixels/sec |
| Feature Matching | 0.92 Gpixels/sec |
| Stereo Matching | 355.9 Gpixels/sec |
| Particle Physics | 9094.4 FPS |
| API | OpenCL |
| G3D Mark Score | 12661 |
| G2D Mark | 779 |
| DirectX 11 | 103 FPS |
| DirectX 12 | 51 FPS |
| GPU Compute | 5074 Ops/s |
| GB6 ML Quantized | 7620 |
| Image Classification (SP) | 4272 |
| Image Segmentation (HP) | 6435 |
| Image Super Resolution (Q) | 9635 |
| Face Detection (HP) | 10867 |
| Pose Estimation (Q) | 41061 |
| Text Classification (SP) | 1737 |
| Machine Translation (HP) | 3301 |
| Object Detection (SP) | 5600 |
| Depth Estimation (Q) | 18914 |
| Style Transfer (SP) | 83427 |
| Framework | ONNX |
| Backend | DirectML |
| GB6 ML Single Precision | 10788 |
| GB6 ML Half Precision | 8753 |
| Pixel Fill Rate | 85 GPixel/s |
| Texture Fill Rate | 170 GTexel/s |
| FLOPS (FP32) | 5.4 TFLOPS |
| Memory Type | GDDR6 |
| Memory Size | 6 GB |
| Memory Clock | 1500 MHz |
| Effective Memory Speed | 12000 Mbps |
| Bus | 192-bit |
| ECC | No |
| Memory Bandwidth | 288 GB/s |
| Interface | PCIe 3.0 x16 |
| TGP | 120 W |
| Manufacturing | TSMC |
| Fabrication Process | 12 nm |
| Die Size | 284 mm² |
| Transistor Count | 6 billion |
| Transistor Density | 21.13 MTr/mm² |
| DirectX | 12 |
| Vulkan | 1.3 |
| OpenGL | 4.6 |
| OpenCL | 3 |
| CUDA | 7.5 |
| Ray Tracing | No |
| DLSS | No |
| DisplayPort | 1.4a |
| Vendor | Nvidia |
| Build | Discrete |
| Released | February 22, 2019 |
| Case | Desktop |
| Purpose | Gaming |
| Segment | Entry-level |
| Architecture | Turing |
| GPU Codename | TU116 |
| Recommended CPU | -Intel Core i5 9500or above |
| Steel Nomad Lite Score | 5944 |
| Time Spy | 6339 |
| Solar Bay | 20534 |
| Port Royal | 1650 |
| Fire Strike | 16653 |
| Wild Life Extreme | 13005 |
| Night Raid | 76361 |
| Base Clock | 1500 MHz |
| Boost Clock | 1770 MHz |
| Shading Units | 1536 |
| Texture Mapping Units (TMUs) | 96 |
| Render Output Units (ROPs) | 48 |
| Compute Units (Pipelines) | 24 |
| Tensor Cores | No |
| Ray-tracing Cores | No |
| L1 Cache | 64KB per cluster |
| L2 Cache | 1536KB shared |
| Instructions Per Cycle | 2 IPC |
84/100