NVIDIA CMP 90HX vs NVIDIA P102-101
What is the difference between NVIDIA CMP 90HX and NVIDIA P102-101. Find out which graphics card has better performance.
Graphics Processor (GPU)
| GA102 | GPU Name | GP102 |
| GA102-100-A1 | Model number | GP102-101-A1 |
| Ampere | Architecture | Pascal |
| Samsung | Foundry | TSMC |
| 8 nm | Process Size | 16 nm |
| 28,300 million | Transistors | 11,800 million |
| 628 mm² | Die Size | 471 mm² |
Graphics Card
| 2021 | Release Date | 2018 |
| Mining GPUs | Family | Mining GPUs |
| Unreleased | Production | Active |
| PCIe 4.0 x16 | Bus Interface | PCIe 3.0 x4 |
Memory
| 10 GB | Memory Size | 10 GB |
| GDDR6X | Memory Type | GDDR5 |
| 320 bit | Memory Bus | 320 bit |
| 760.3 GB/s | Bandwidth | 320.3 GB/s |
Performance
| 136.8 GPixel/s | Pixel fillrate | 133.6 GPixel/s |
| 342.0 GTexel/s | Texture fillrate | 334.0 GTexel/s |
| 21.89 TFLOPS (1:1) | FP16 (half) performance | 167.0 GFLOPS (1:64) |
| 21.89 TFLOPS | FP32 (float) performance | 10.69 TFLOPS |
| 342.0 GFLOPS (1:64) | FP64 (double) performance | 334.0 GFLOPS (1:32) |
Clock Speeds
| 1440 MHz | Base Clock | 1557 MHz |
| 1710 MHz | Boost Clock | 1670 MHz |
| 1188 MHz 19 Gbps effective | Memory Clock | 2002 MHz 8 Gbps effective |
Render Config
| 6400 | Shading Units | 3200 |
| 200 | TMUs | 200 |
| 80 | ROPs | 80 |
| 128 KB (per SM) | L1 Cache | 48 KB (per SM) |
| 5 MB | L2 Cache | 2.5 MB |
| 50 | SM Count | 25 |
Board Design
| Dual-slot | Slot Width | Dual-slot |
| 285 mm 11.2 inches | Length | 267 mm 10.5 inches |
| 320 W | Thermal design power (TDP) | 250 W |
| 700 W | Suggested PSU | 600 W |
| No outputs | Display Connectors | No outputs |
| 2x 8-pin | Power Connectors | 2x 8-pin |
| PG132 SKU 100 | Board Number | PG611 SKU 101 |
API support
| 12 Ultimate (12_2) | DirectX | 12 (12_1) |
| 4.6 | OpenGL | 4.6 |
| 3.0 | OpenCL | 3.0 |
| 1.2 | Vulkan | 1.2 |
| 6.6 | Shader Model | 6.4 |
| 8.6 | CUDA | 6.1 |