Tesla P40 and Tesla K80m - video card comparison
Tesla P40
Tesla K80m
Architecture
Pascal (2016−2021)
Kepler 2.0 (2013−2015)
GPU code name
GP102
GK210
Market segment
Workstation
Workstation
Release date
September 2016
November 2014
Interface
PCIe 3.0 x16
PCIe 3.0 x16
Supplementary power connectors
8-pin EPS
1x 8-pin
Boost clock speed
1531 MHz
824 MHz
Core clock speed
1303 MHz
562 MHz
Floating-point performance
11,758 gflops
2x 4,373 gflops
Manufacturing process technology
16 nm
28 nm
Number of transistors
11,800 million
7,100 million
Pipelines / CUDA cores
3840
2x 2496
Power consumption (TDP)
250 Watt
300 Watt
Texture fill rate
367.4
2x 182.2 billion/sec
Memory type
GDDR5
GDDR5
Maximum RAM amount
24 GB
2x 12 GB
Memory bus width
384 Bit
2x 384 Bit
Memory clock speed
7.2 GB/s
5012 MHz
Memory bandwidth
347.1 GB/s
2x 240.6 GB/s
Display Connectors
No outputs
No outputs
DirectX
12 (12_1)
12.0 (11_1)
CUDA
6.1
-
OpenCL
3.0
-
OpenGL
4.6
4.6
Shader Model
6.7
-
Vulkan
1.3
-