Quadro K620 and Tesla M2070 - video card comparison
Quadro K620
Tesla M2070
Architecture
Maxwell (2014−2018)
Fermi (2010−2014)
GPU code name
GM107
GF100
Market segment
Workstation
Workstation
Release date
July 2014
July 2011
Interface
PCIe 2.0 x16
PCIe 2.0 x16
Supplementary power connectors
None
1x 6-pin + 1x 8-pin
Boost clock speed
1124 MHz
-
Core clock speed
1058 MHz
574 MHz
Floating-point performance
863.2 gflops
1,030.4 gflops
Manufacturing process technology
28 nm
40 nm
Number of transistors
1,870 million
3,100 million
Pipelines / CUDA cores
384
448
Power consumption (TDP)
41 Watt
225 Watt
Texture fill rate
26.98
32.14
Memory type
128 Bit
GDDR5
Maximum RAM amount
2 GB
6 GB
Memory bus width
128 Bit
384 Bit
Memory clock speed
1800 MHz
3132 MHz
Memory bandwidth
Up to 29 GB/s
150.3 GB/s
Display Connectors
DVI-I DP
No outputs
Number of simultaneous displays
4
-
DirectX
12
12 (11_0)
CUDA
5.0
2.0
OpenCL
1.2
1.1
OpenGL
4.5
4.6
Shader Model
5
5.1
Vulkan
+
N/A
3D Vision Pro
+
-
Mosaic
+
-
nView Desktop Management
+
-