Tesla M40 and Tesla K80m - video card comparison
Tesla M40
Tesla K80m
Architecture
Maxwell 2.0 (2015−2019)
Kepler 2.0 (2013−2015)
GPU code name
GM200
GK210
Market segment
Workstation
Workstation
Release date
November 2015
November 2014
Interface
PCIe 3.0 x16
PCIe 3.0 x16
Supplementary power connectors
8-pin EPS
1x 8-pin
Boost clock speed
1112 MHz
824 MHz
Core clock speed
948 MHz
562 MHz
Floating-point performance
6,844 gflops
2x 4,373 gflops
Manufacturing process technology
28 nm
28 nm
Number of transistors
8,000 million
7,100 million
Pipelines / CUDA cores
3072
2x 2496
Power consumption (TDP)
250 Watt
300 Watt
Texture fill rate
213.5
2x 182.2 billion/sec
Memory type
GDDR5
GDDR5
Maximum RAM amount
12 GB
2x 12 GB
Memory bus width
384 Bit
2x 384 Bit
Memory clock speed
6 GB/s
5012 MHz
Memory bandwidth
288.4 GB/s
2x 240.6 GB/s
Display Connectors
No outputs
No outputs
DirectX
12 (12_1)
12.0 (11_1)
CUDA
5.2
-
OpenCL
3.0
-
OpenGL
4.6
4.6
Shader Model
6.7
-
Vulkan
1.3
-