NVIDIA GeForce GTX 1660 Ti vs NVIDIA Tesla K20X
What is the difference between NVIDIA GeForce GTX 1660 Ti and NVIDIA Tesla K20X. Find out which graphics card has better performance.
Graphics Processor (GPU)
| TU116 | GPU Name | GK110 |
| Turing | Architecture | Kepler |
| TSMC | Foundry | TSMC |
| 12 nm | Process Size | 28 nm |
| 6,600 million | Transistors | 7,080 million |
| 284 mm² | Die Size | 561 mm² |
Graphics Card
| Feb 22nd, 2019 | Release Date | Nov 12th, 2012 |
| GeForce 16 | Family | Tesla (Kxx) |
| Active | Production | End-of-life |
| 279 USD | Launch Price | 7,699 USD |
| PCIe 3.0 x16 | Bus Interface | PCIe 3.0 x16 |
Memory
| 6 GB | Memory Size | 6 GB |
| GDDR6 | Memory Type | GDDR5 |
| 192 bit | Memory Bus | 384 bit |
| 288.0 GB/s | Bandwidth | 249.6 GB/s |
Performance
| 84.96 GPixel/s | Pixel fillrate | 40.99 GPixel/s |
| 169.9 GTexel/s | Texture fillrate | 164.0 GTexel/s |
| 5.437 TFLOPS | FP32 (float) performance | 3.935 TFLOPS |
| 169.9 GFLOPS (1:32) | FP64 (double) performance | 1,312 GFLOPS (1:3) |
Clock Speeds
| 1500 MHz 12 Gbps effective | Memory Clock | 1300 MHz 5.2 Gbps effective |
Render Config
| 1536 | Shading Units | 2688 |
| 96 | TMUs | 224 |
| 48 | ROPs | 48 |
| 64 KB (per SM) | L1 Cache | 16 KB (per SMX) |
| 1536 KB | L2 Cache | 1536 KB |
Board Design
| Dual-slot | Slot Width | Dual-slot |
| 229 mm 9 inches | Length | 267 mm 10.5 inches |
| 120 W | Thermal design power (TDP) | 235 W |
| 300 W | Suggested PSU | 550 W |
| 1x DVI 1x HDMI 1x DisplayPort | Display Connectors | No outputs |
| 1x 8-pin | Power Connectors | 1x 6-pin + 1x 8-pin |
| PG161 SKU 21 | Board Number | P2081 |
API support
| 12 (12_1) | DirectX | 12 (11_0) |
| 4.6 | OpenGL | 4.6 |
| 3.0 | OpenCL | 3.0 |
| 1.2 | Vulkan | 1.1 |
| 6.6 | Shader Model | 5.1 |
| 7.5 | CUDA | 3.5 |