NVIDIA P102-100 vs NVIDIA P104-100
Graphics Processor
| GPU model | GP102 | GP104 |
|---|---|---|
| GPU variant | GP102-100-A1 | GP104-100-A1 |
| Architecture | Pascal | Pascal |
| Foundry | TSMC | TSMC |
| Manufacturing process | 16 nm | 16 nm |
| Number of transistors | 11,800 million | 7,200 million |
| Die size | 471 mm² | 314 mm² |
Graphics Card
| Release date | Feb 12th, 2018 | Dec 12th, 2017 |
|---|---|---|
| Generation | Mining GPUs | Mining GPUs |
| Production | End-of-life | End-of-life |
| Interface | PCIe 3.0 x4 | PCIe 3.0 x16 |
Clocks
| Base clock | 1582 MHz | 1607 MHz |
|---|---|---|
| Boost clock | 1683 MHz | 1733 MHz |
| Memory clock | 1376 MHz 11 Gbps effective | 1251 MHz 10 Gbps effective |
Memory Configuration
| Memory size | 5 GB | 4 GB |
|---|---|---|
| Memory type | GDDR5X | GDDR5X |
| Memory bus width | 320 bit | 256 bit |
| Bandwidth | 440.3 GB/s | 320.3 GB/s |
Render Configuration
| Shading units | 3200 | 1920 |
|---|---|---|
| TMUs | 200 | 120 |
| ROPs | 80 | 64 |
| SM count | 25 | 15 |
| Cache L1 | 48 KB (per SM) | 48 KB (per SM) |
| Cache L2 | 2.5 MB | 2 MB |
Performance
| Pixel rate | 134.6 GPixel/s | 110.9 GPixel/s |
|---|---|---|
| Texture rate | 336.6 GTexel/s | 208.0 GTexel/s |
| FP16 (half) performance | 168.3 GFLOPS (1:64) | 104.0 GFLOPS (1:64) |
| FP32 (float) performance | 10.77 TFLOPS | 6.655 TFLOPS |
| FP64 (double) performance | 336.6 GFLOPS (1:32) | 208.0 GFLOPS (1:32) |
Dimensions & Outputs
| Slot width | Dual-slot | Dual-slot |
|---|---|---|
| Length | 267 mm 10.5 inches | 267 mm 10.5 inches |
| TDP | 250 W | unknown |
| Suggested PSU | 600 W | 200 W |
| Outputs | No outputs | No outputs |
| Power connectors | 2x 8-pin | 1x 8-pin |
| Board number | PG611 SKU 100 | PG413 SKU 100 |
API Support & Features
| DirectX | 12 (12_1) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| CUDA | 6.1 | 6.1 |
| Shader model | 6.4 | 6.4 |
