NVIDIA Tesla K80 vs NVIDIA Tesla M40

Comparative analysis of NVIDIA Tesla K80 and NVIDIA Tesla M40 videocards for all known characteristics in the following categories: Essentials, Technical info, Video outputs and ports, Compatibility, dimensions and requirements, API support, Memory. Benchmark videocards performance analysis: PassMark - G2D Mark, PassMark - G3D Mark, Geekbench - OpenCL, CompuBench 1.5 Desktop - Face Detection (mPixels/s), CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s), CompuBench 1.5 Desktop - T-Rex (Frames/s), CompuBench 1.5 Desktop - Video Composition (Frames/s), CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s).

NVIDIA Tesla K80

NVIDIA Tesla M40

Differences

Reasons to consider the NVIDIA Tesla K80

801.3x more texture fill rate: 171.4 GTexel/s vs 213.9 GTexel / s

Texture fill rate	171.4 GTexel/s vs 213.9 GTexel / s

Reasons to consider the NVIDIA Tesla M40

Videocard is newer: launch date 11 month(s) later
Around 69% higher core clock speed: 948 MHz vs 562 MHz
Around 35% higher boost clock speed: 1114 MHz vs 824 MHz
Around 23% higher pipelines: 3072 vs 2496
Around 20% lower typical power consumption: 250 Watt vs 300 Watt
4.8x more memory clock speed: 6008 MHz vs 1253 MHz, 5 Gbps effective
Around 19% better performance in PassMark - G2D Mark: 445 vs 374
Around 79% better performance in PassMark - G3D Mark: 10448 vs 5852

Specifications (specs)
Launch date	10 November 2015 vs 17 Nov 2014
Core clock speed	948 MHz vs 562 MHz
Boost clock speed	1114 MHz vs 824 MHz
Pipelines	3072 vs 2496
Thermal Design Power (TDP)	250 Watt vs 300 Watt
Memory clock speed	6008 MHz vs 1253 MHz, 5 Gbps effective
Benchmarks
PassMark - G2D Mark	445 vs 374
PassMark - G3D Mark	10448 vs 5852

Compare benchmarks

GPU 1: NVIDIA Tesla K80
GPU 2: NVIDIA Tesla M40

PassMark - G2D Mark

GPU 1

GPU 2

374

445

PassMark - G3D Mark

GPU 1

GPU 2

5852

10448

Name	NVIDIA Tesla K80	NVIDIA Tesla M40
PassMark - G2D Mark	374	445
PassMark - G3D Mark	5852	10448
Geekbench - OpenCL		39179
CompuBench 1.5 Desktop - Face Detection (mPixels/s)		183.81
CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s)		2637.997
CompuBench 1.5 Desktop - T-Rex (Frames/s)		13.059
CompuBench 1.5 Desktop - Video Composition (Frames/s)		160.359
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s)		688.388

Compare specifications (specs)

	NVIDIA Tesla K80	NVIDIA Tesla M40
Essentials
Architecture	Kepler 2.0	Maxwell 2.0
Code name	GK210	GM200
Launch date	17 Nov 2014	10 November 2015
Place in performance rating	256	257
Type		Workstation
Technical info
Boost clock speed	824 MHz	1114 MHz
Core clock speed	562 MHz	948 MHz
Manufacturing process technology	28 nm	28 nm
Peak Double Precision (FP64) Performance	1,371 GFLOPS (1:3)
Peak Single Precision (FP32) Performance	4.113 TFLOPS
Pipelines	2496	3072
Pixel fill rate	42.85 GPixel/s
Texture fill rate	171.4 GTexel/s	213.9 GTexel / s
Thermal Design Power (TDP)	300 Watt	250 Watt
Transistor count	7100 million	8,000 million
Floating-point performance		6,844 gflops
Video outputs and ports
Display Connectors	No outputs	No outputs
Compatibility, dimensions and requirements
Form factor	Dual-slot
Interface	PCIe 3.0 x16	PCIe 3.0 x16
Length	267 mm, 10.5 inches	267 mm
Recommended system power (PSU)	700 Watt
Supplementary power connectors	1x 8-pin	1x 6-pin + 1x 8-pin
API support
DirectX	12 (11_1)	12.0 (12_1)
OpenCL	3.0
OpenGL	4.6	4.6
Shader Model	6.5 (5.1)
Vulkan
Memory
Maximum RAM amount	12 GB	12 GB
Memory bandwidth	240.6 GB/s	288.0 GB / s
Memory bus width	384 bit	384 Bit
Memory clock speed	1253 MHz, 5 Gbps effective	6008 MHz
Memory type	GDDR5	GDDR5