NVIDIA Tesla K20Xm vs NVIDIA GeForce GTX 660 Ti

Comparative analysis of NVIDIA Tesla K20Xm and NVIDIA GeForce GTX 660 Ti videocards for all known characteristics in the following categories: Essentials, Technical info, Video outputs and ports, Compatibility, dimensions and requirements, API support, Memory, Technologies. 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 - Manhattan (Frames), GFXBench 4.0 - T-Rex (Frames), GFXBench 4.0 - Manhattan (Fps), GFXBench 4.0 - T-Rex (Fps), GFXBench 4.0 - Car Chase Offscreen (Frames), GFXBench 4.0 - Car Chase Offscreen (Fps), 3DMark Fire Strike - Graphics Score.

 

Differences

Reasons to consider the NVIDIA Tesla K20Xm

  • Videocard is newer: launch date 2 month(s) later
  • Around 60% higher texture fill rate: 164.0 GTexel / s vs 102.5 billion / sec
  • 2x more pipelines: 2688 vs 1344
  • Around 60% better floating-point performance: 3,935 gflops vs 2,459.5 gflops
  • 3x more maximum memory size: 6 GB vs 2 GB
  • 866.7x more memory clock speed: 5200 MHz vs 6.0 GB/s
  • Around 9% better performance in Geekbench - OpenCL: 16664 vs 15347
  • Around 36% better performance in CompuBench 1.5 Desktop - Face Detection (mPixels/s): 49.455 vs 36.256
  • Around 62% better performance in CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s): 1666.354 vs 1029.267
  • Around 92% better performance in CompuBench 1.5 Desktop - T-Rex (Frames/s): 5.836 vs 3.037
  • 2x better performance in CompuBench 1.5 Desktop - Video Composition (Frames/s): 92.472 vs 45.894
  • 3.3x better performance in CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s): 275.229 vs 82.191
  • 2.1x better performance in GFXBench 4.0 - T-Rex (Frames): 7038 vs 3337
  • 2.1x better performance in GFXBench 4.0 - T-Rex (Fps): 7038 vs 3337
Specifications (specs)
Launch date 12 November 2012 vs 16 August 2012
Texture fill rate 164.0 GTexel / s vs 102.5 billion / sec
Pipelines 2688 vs 1344
Floating-point performance 3,935 gflops vs 2,459.5 gflops
Maximum memory size 6 GB vs 2 GB
Memory clock speed 5200 MHz vs 6.0 GB/s
Benchmarks
Geekbench - OpenCL 16664 vs 15347
CompuBench 1.5 Desktop - Face Detection (mPixels/s) 49.455 vs 36.256
CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s) 1666.354 vs 1029.267
CompuBench 1.5 Desktop - T-Rex (Frames/s) 5.836 vs 3.037
CompuBench 1.5 Desktop - Video Composition (Frames/s) 92.472 vs 45.894
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s) 275.229 vs 82.191
GFXBench 4.0 - T-Rex (Frames) 7038 vs 3337
GFXBench 4.0 - T-Rex (Fps) 7038 vs 3337

Reasons to consider the NVIDIA GeForce GTX 660 Ti

  • Around 25% higher core clock speed: 915 MHz vs 732 MHz
  • Around 57% lower typical power consumption: 150 Watt vs 235 Watt
  • 2.4x better performance in PassMark - G2D Mark: 466 vs 198
  • 3.6x better performance in GFXBench 4.0 - Manhattan (Frames): 3684 vs 1032
  • 3.6x better performance in GFXBench 4.0 - Manhattan (Fps): 3684 vs 1032
Specifications (specs)
Core clock speed 915 MHz vs 732 MHz
Thermal Design Power (TDP) 150 Watt vs 235 Watt
Benchmarks
PassMark - G3D Mark 4423 vs 4403
PassMark - G2D Mark 466 vs 198
GFXBench 4.0 - Manhattan (Frames) 3684 vs 1032
GFXBench 4.0 - Manhattan (Fps) 3684 vs 1032

Compare benchmarks

GPU 1: NVIDIA Tesla K20Xm
GPU 2: NVIDIA GeForce GTX 660 Ti

PassMark - G3D Mark
GPU 1
GPU 2
4403
4423
PassMark - G2D Mark
GPU 1
GPU 2
198
466
Geekbench - OpenCL
GPU 1
GPU 2
16664
15347
CompuBench 1.5 Desktop - Face Detection (mPixels/s)
GPU 1
GPU 2
49.455
36.256
CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s)
GPU 1
GPU 2
1666.354
1029.267
CompuBench 1.5 Desktop - T-Rex (Frames/s)
GPU 1
GPU 2
5.836
3.037
CompuBench 1.5 Desktop - Video Composition (Frames/s)
GPU 1
GPU 2
92.472
45.894
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s)
GPU 1
GPU 2
275.229
82.191
GFXBench 4.0 - Manhattan (Frames)
GPU 1
GPU 2
1032
3684
GFXBench 4.0 - T-Rex (Frames)
GPU 1
GPU 2
7038
3337
GFXBench 4.0 - Manhattan (Fps)
GPU 1
GPU 2
1032
3684
GFXBench 4.0 - T-Rex (Fps)
GPU 1
GPU 2
7038
3337
Name NVIDIA Tesla K20Xm NVIDIA GeForce GTX 660 Ti
PassMark - G3D Mark 4403 4423
PassMark - G2D Mark 198 466
Geekbench - OpenCL 16664 15347
CompuBench 1.5 Desktop - Face Detection (mPixels/s) 49.455 36.256
CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s) 1666.354 1029.267
CompuBench 1.5 Desktop - T-Rex (Frames/s) 5.836 3.037
CompuBench 1.5 Desktop - Video Composition (Frames/s) 92.472 45.894
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s) 275.229 82.191
GFXBench 4.0 - Manhattan (Frames) 1032 3684
GFXBench 4.0 - T-Rex (Frames) 7038 3337
GFXBench 4.0 - Manhattan (Fps) 1032 3684
GFXBench 4.0 - T-Rex (Fps) 7038 3337
GFXBench 4.0 - Car Chase Offscreen (Frames) 3285
GFXBench 4.0 - Car Chase Offscreen (Fps) 3285
3DMark Fire Strike - Graphics Score 1616

Compare specifications (specs)

NVIDIA Tesla K20Xm NVIDIA GeForce GTX 660 Ti

Essentials

Architecture Kepler Kepler
Code name GK110 GK104
Launch date 12 November 2012 16 August 2012
Launch price (MSRP) $7,699 $299
Place in performance rating 746 747
Type Workstation Desktop
Price now $321.99
Value for money (0-100) 17.34

Technical info

Core clock speed 732 MHz 915 MHz
Floating-point performance 3,935 gflops 2,459.5 gflops
Manufacturing process technology 28 nm 28 nm
Pipelines 2688 1344
Texture fill rate 164.0 GTexel / s 102.5 billion / sec
Thermal Design Power (TDP) 235 Watt 150 Watt
Transistor count 7,080 million 3,540 million
Boost clock speed 980 MHz
CUDA cores 1344

Video outputs and ports

Display Connectors No outputs 2x DVI, 1x HDMI, 1x DisplayPort, One Dual Link DVI-I, One Dual Link DVI-D, One HDMI...
Audio input for HDMI Internal
G-SYNC support
HDCP
HDMI
Maximum VGA resolution 2048x1536
Multi monitor support

Compatibility, dimensions and requirements

Interface PCIe 3.0 x16 PCIe 3.0 x16
Length 267 mm 9.5" (24.1 cm)
Bus support PCI Express 3.0
Height 4.376" (11.1 cm)
SLI options 3-way
Supplementary power connectors Two 6-pin

API support

DirectX 12.0 (11_0) 12.0 (11_0)
OpenGL 4.6 4.3

Memory

Maximum RAM amount 6 GB 2 GB
Memory bandwidth 249.6 GB / s 144.2 GB / s
Memory bus width 384 Bit 192-bit GDDR5
Memory clock speed 5200 MHz 6.0 GB/s
Memory type GDDR5 GDDR5
Shared memory 0

Technologies

3D Blu-Ray
3D Gaming
3D Vision
Adaptive VSync
CUDA
FXAA
GPU Boost
SLI
TXAA