NVIDIA GRID K2 vs NVIDIA Tesla K20c

Comparative analysis of NVIDIA GRID K2 and NVIDIA Tesla K20c 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 - G3D Mark, PassMark - G2D 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), GFXBench 4.0 - Car Chase Offscreen (Frames), GFXBench 4.0 - Manhattan (Frames), GFXBench 4.0 - T-Rex (Frames), GFXBench 4.0 - Car Chase Offscreen (Fps), GFXBench 4.0 - Manhattan (Fps), GFXBench 4.0 - T-Rex (Fps).

NVIDIA GRID K2

NVIDIA Tesla K20c

Differences

Reasons to consider the NVIDIA GRID K2

Videocard is newer: launch date 5 month(s) later
Around 6% higher core clock speed: 745 MHz vs 706 MHz
2x more texture fill rate: 2x 95.36 GTexel / s billion / sec vs 146.8 GTexel / s
Around 23% higher pipelines: 2x 1536 vs 2496
Around 30% better floating-point performance: 2x 2,289 gflops vs 3,524 gflops
Around 60% higher maximum memory size: 2x 4 GB vs 5 GB

Launch date	11 May 2013 vs 12 November 2012
Core clock speed	745 MHz vs 706 MHz
Texture fill rate	2x 95.36 GTexel / s billion / sec vs 146.8 GTexel / s
Pipelines	2x 1536 vs 2496
Floating-point performance	2x 2,289 gflops vs 3,524 gflops
Maximum memory size	2x 4 GB vs 5 GB

Reasons to consider the NVIDIA Tesla K20c

Around 4% higher memory clock speed: 5200 MHz vs 5000 MHz
Around 49% better performance in Geekbench - OpenCL: 15597 vs 10471
2.1x better performance in CompuBench 1.5 Desktop - Face Detection (mPixels/s): 39.423 vs 18.943
Around 47% better performance in CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s): 1427.15 vs 968.568
Around 64% better performance in CompuBench 1.5 Desktop - T-Rex (Frames/s): 4.128 vs 2.51
Around 95% better performance in CompuBench 1.5 Desktop - Video Composition (Frames/s): 64.383 vs 32.988
Around 57% better performance in CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s): 179.044 vs 114.144

Specifications (specs)
Memory clock speed	5200 MHz vs 5000 MHz
Benchmarks
Geekbench - OpenCL	15597 vs 10471
CompuBench 1.5 Desktop - Face Detection (mPixels/s)	39.423 vs 18.943
CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s)	1427.15 vs 968.568
CompuBench 1.5 Desktop - T-Rex (Frames/s)	4.128 vs 2.51
CompuBench 1.5 Desktop - Video Composition (Frames/s)	64.383 vs 32.988
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s)	179.044 vs 114.144

Compare benchmarks

GPU 1: NVIDIA GRID K2
GPU 2: NVIDIA Tesla K20c

Geekbench - OpenCL

GPU 1

GPU 2

10471

15597

CompuBench 1.5 Desktop - Face Detection (mPixels/s)

GPU 1

GPU 2

18.943

39.423

CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s)

GPU 1

GPU 2

968.568

1427.15

CompuBench 1.5 Desktop - T-Rex (Frames/s)

GPU 1

GPU 2

2.51

4.128

CompuBench 1.5 Desktop - Video Composition (Frames/s)

GPU 1

GPU 2

32.988

64.383

CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s)

GPU 1

GPU 2

114.144

179.044

Name	NVIDIA GRID K2	NVIDIA Tesla K20c
PassMark - G3D Mark	2737
PassMark - G2D Mark	319
Geekbench - OpenCL	10471	15597
CompuBench 1.5 Desktop - Face Detection (mPixels/s)	18.943	39.423
CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s)	968.568	1427.15
CompuBench 1.5 Desktop - T-Rex (Frames/s)	2.51	4.128
CompuBench 1.5 Desktop - Video Composition (Frames/s)	32.988	64.383
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s)	114.144	179.044
GFXBench 4.0 - Car Chase Offscreen (Frames)	6371
GFXBench 4.0 - Manhattan (Frames)	344
GFXBench 4.0 - T-Rex (Frames)	312
GFXBench 4.0 - Car Chase Offscreen (Fps)	6371
GFXBench 4.0 - Manhattan (Fps)	344
GFXBench 4.0 - T-Rex (Fps)	312

Compare specifications (specs)

	NVIDIA GRID K2	NVIDIA Tesla K20c
Essentials
Architecture	Kepler	Kepler
Code name	GK104	GK110
Launch date	11 May 2013	12 November 2012
Launch price (MSRP)	$5,199	$3,199
Place in performance rating	975	976
Type	Workstation	Workstation
Technical info
Core clock speed	745 MHz	706 MHz
Floating-point performance	2x 2,289 gflops	3,524 gflops
Manufacturing process technology	28 nm	28 nm
Pipelines	2x 1536	2496
Texture fill rate	2x 95.36 GTexel / s billion / sec	146.8 GTexel / s
Thermal Design Power (TDP)	225 Watt	225 Watt
Transistor count	3,540 million	7,080 million
Video outputs and ports
Display Connectors	No outputs	No outputs
Compatibility, dimensions and requirements
Interface	PCIe 3.0 x16	PCIe 2.0 x16
Length	267 mm	267 mm
Supplementary power connectors	1x 8-pin	1x 6-pin + 1x 8-pin
API support
DirectX	12.0 (11_0)	12.0 (11_0)
OpenGL	4.6	4.6
Memory
Maximum RAM amount	2x 4 GB	5 GB
Memory bandwidth	2x 160.0 GB / s	208.0 GB / s
Memory bus width	2x 256 Bit	320 Bit
Memory clock speed	5000 MHz	5200 MHz
Memory type	GDDR5	GDDR5