NVIDIA Jetson TK1 GPU vs NVIDIA P104-100
Graphics Processor
| GPU model | GK20A | GP104 |
|---|---|---|
| Architecture | Kepler 2.0 | Pascal |
| Foundry | TSMC | TSMC |
| Manufacturing process | 28 nm | 16 nm |
| Number of transistors | unknown | 7,200 million |
| Die size | unknown | 314 mm² |
| GPU variant | — | GP104-100-A1 |
Integrated Graphics
| Release date | Oct 15th, 2014 | — |
|---|---|---|
| Generation | Tegra | — |
| Production | End-of-life | — |
| Interface | IGP | — |
Clocks
| Base clock | 756 MHz | 1607 MHz |
|---|---|---|
| Boost clock | 951 MHz | 1733 MHz |
| Memory clock | System Shared | 1251 MHz 10 Gbps effective |
Memory Configuration
| Memory size | System Shared | 4 GB |
|---|---|---|
| Memory type | System Shared | GDDR5X |
| Memory bus width | System Shared | 256 bit |
| Bandwidth | System Dependent | 320.3 GB/s |
Render Configuration
| Shading units | 192 | 1920 |
|---|---|---|
| TMUs | 8 | 120 |
| ROPs | 4 | 64 |
| SMX count | 1 | — |
| Cache L1 | 16 KB (per SMX) | 48 KB (per SM) |
| Cache L2 | 128 KB | 2 MB |
| SM count | — | 15 |
Performance
| Pixel rate | 3.804 GPixel/s | 110.9 GPixel/s |
|---|---|---|
| Texture rate | 7.608 GTexel/s | 208.0 GTexel/s |
| FP32 (float) performance | 365.2 GFLOPS | 6.655 TFLOPS |
| FP64 (double) performance | 11.41 GFLOPS (1:32) | 208.0 GFLOPS (1:32) |
| FP16 (half) performance | — | 104.0 GFLOPS (1:64) |
Dimensions & Outputs
| Slot width | IGP | Dual-slot |
|---|---|---|
| TDP | 8 W | unknown |
| Outputs | No outputs | No outputs |
| Length | — | 267 mm 10.5 inches |
| Suggested PSU | — | 200 W |
| Power connectors | — | 1x 8-pin |
| Board number | — | PG413 SKU 100 |
API Support & Features
| DirectX | 12 (11_0) | 12 (12_1) |
|---|---|---|
| OpenGL | ES 3.1 | 4.6 |
| OpenCL | — | 3.0 |
| Vulkan | 1.1 | 1.2 |
| CUDA | 3.2 | 6.1 |
| Shader model | 5.1 | 6.4 |
Other Features
Graphics Card
| Release date | — | Dec 12th, 2017 |
|---|---|---|
| Generation | — | Mining GPUs |
| Production | — | End-of-life |
| Interface | — | PCIe 3.0 x16 |
