Home NVIDIA P102 100

NVIDIA P102 100

This is a GPU manufactured with TSMC 16nm process, based on Nvidia Pascal architecture and released on Feb 2018. It features 11.8 billion transistors, 3200 CUDA cores and 5GB GDDR5X memory, theoretical performance of 10.77TFLOPS, with total power consumption of 250W.

Graphics Card

[Report Issues]
Release Date
Feb 2018
Generation
Mining GPUs
Type
Desktop
Bus Interface
PCIe 3.0 x4

Clock Speeds

[Report Issues]
Base Clock
1582MHz
Boost Clock
1683MHz
Memory Clock
1376MHz

Memory

[Report Issues]
Memory Size
5GB
Memory Type
GDDR5X
Memory Bus
320bit
Bandwidth
440.3 GB/s

Render Config

[Report Issues]
SM Count
25
Shading Units
3200
TMUs
200
ROPs
80
L1 Cache
48 KB (per SM)
L2 Cache
0MB

Theoretical Performance

[Report Issues]
Pixel Rate
134.6 GPixel/s
Texture Rate
336.6 GTexel/s
FP16 (half)
168.3 GFLOPS
FP32 (float)
10.77 TFLOPS
FP64 (double)
336.6 GFLOPS

Graphics Processor

[Report Issues]
GPU Name
GP102
GPU Variant
GP102-100-A1
Architecture
Pascal
Foundry
TSMC
Process Size
16 nm
Transistors
11,800 million
Die Size
471mm²

Board Design

[Report Issues]
TDP
250W
Suggested PSU
600W
Outputs
No outputs
Power Connectors
2x 8-pin

Graphics Features

[Report Issues]
DirectX
12 (12_1)
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
CUDA
6.1
Shader Model
6.4

Rankings

[Report Issues]
FP32 (float)
10.94 TFLOPS
10.87 TFLOPS
10.79 TFLOPS
NVIDIA P102 100 5 GB GDDR5X
10.77 TFLOPS
10.75 TFLOPS
10.75 TFLOPS
10.75 TFLOPS
Blender
872
NVIDIA P102 100 5 GB GDDR5X
844
NVIDIA Quadro P6000 24 GB GDDR5X
842
824

Related Comparisons

© 2024 - TopCPU.net   Contact Us Privacy Policy