GPU model | GP102 | NV17 |
---|---|---|
GPU variant | GP102-100-A1 | NV17 GL |
Architecture | Pascal | Celsius |
Foundry | TSMC | TSMC |
Manufacturing process | 16 nm | 150 nm |
Number of transistors | 11,800 million | 29 million |
Die size | 471 mm² | 65 mm² |
Release date | Feb 12th, 2018 | Feb 19th, 2002 |
---|---|---|
Generation | Mining GPUs | Quadro4 XGL |
Production | End-of-life | End-of-life |
Interface | PCIe 3.0 x4 | AGP 4x |
Reviews | — | 26 in our database |
Base clock | 1582 MHz | — |
---|---|---|
Boost clock | 1683 MHz | — |
Memory clock | 1376 MHz 11 Gbps effective | 166 MHz |
GPU clock | — | 250 MHz |
Memory size | 5 GB | 128 MB |
---|---|---|
Memory type | GDDR5X | SDR |
Memory bus width | 320 bit | 128 bit |
Bandwidth | 440.3 GB/s | 2.656 GB/s |
Shading units | 3200 | — |
---|---|---|
TMUs | 200 | 4 |
ROPs | 80 | 4 |
SM count | 25 | — |
Cache L1 | 48 KB (per SM) | — |
Cache L2 | 2.5 MB | — |
Pixel shaders | — | 2 |
Vertex shaders | — | 0 |
Pixel rate | 134.6 GPixel/s | 1.000 GPixel/s |
---|---|---|
Texture rate | 336.6 GTexel/s | 1.000 GTexel/s |
FP16 (half) performance | 168.3 GFLOPS (1:64) | — |
FP32 (float) performance | 10.77 TFLOPS | — |
FP64 (double) performance | 336.6 GFLOPS (1:32) | — |
Slot width | Dual-slot | Single-slot |
---|---|---|
Length | 267 mm 10.5 inches | 168 mm 6.6 inches |
TDP | 250 W | unknown |
Suggested PSU | 600 W | 200 W |
Outputs | No outputs | 1x LFH60 |
Power connectors | 2x 8-pin | None |
Board number | PG611 SKU 100 | — |
DirectX | 12 (12_1) | 7.0 |
---|---|---|
OpenGL | 4.6 | 1.5 |
OpenCL | 3.0 | — |
Vulkan | 1.2 | — |
CUDA | 6.1 | — |
Shader model | 6.4 | — |
Pixel shader | — | 1.1 |
Vertex shader | — | 1.1 |