Report an Error

NVIDIA Quadro K620

GM107
Graphics Processor
384
Cores
16
TMUs
16
ROPs
2 GB
Memory Size
DDR3
Memory Type
128 bit
Bus Width
NVIDIA Quadro K620 Photo NVIDIA GM107 Photo
The Quadro K620 was a professional graphics card by NVIDIA, launched in July 2014. Built on the 28 nm process, and based on the GM107 graphics processor, in its GM107-850-A2 variant, the card supports DirectX 12.0. The GM107 graphics processor is an average sized chip with a die area of 148 mm² and 1,870 million transistors. Unlike the fully unlocked GeForce GTX 750 Ti, which uses the same GPU but has all 640 shaders enabled, NVIDIA has disabled some shading units on the Quadro K620 to reach the product's target shader count. It features 384 shading units, 16 texture mapping units and 16 ROPs. NVIDIA has placed 2,048 MB DDR3 memory on the card, which are connected using a 128-bit memory interface. The GPU is operating at a frequency of 1058 MHz, which can be boosted up to 1124 MHz, memory is running at 900 MHz.
Being a single-slot card, the NVIDIA Quadro K620 does not require any additional power connector, its power draw is rated at 41 W maximum. Display outputs include: 1x DVI, 1x DisplayPort. Quadro K620 is connected to the rest of the system using a PCI-Express 2.0 x16 interface. The card measures 160 mm in length, and features a single-slot cooling solution.

Graphics Processor

GPU Name
GM107
GPU Variant
GM107-850-A2
Architecture
Maxwell
Foundry
TSMC
Process Size
28 nm
Transistors
1,870 million
Die Size
148 mm²

Graphics Card

Release Date
Jul 22nd, 2014
Generation
Quadro
(Kx200)
Production
End-of-life
Bus Interface
PCIe 2.0 x16

Relative Performance

9%
11%
14%
18%
21%
22%
27%
36%
36%
39%
42%
42%
49%
52%
56%
57%
60%
62%
66%
69%
71%
73%
76%
86%
87%
89%
90%
90%
91%
92%
97%
100%
Quadro K620
102%
104%
107%
108%
109%
110%
111%
111%
113%
118%
121%
130%
132%
132%
138%
138%
143%
152%
157%
158%
161%
162%
167%
169%
171%
174%
182%
185%
188%
188%
189%
189%
204%
206%
211%
214%
215%
216%
217%
221%
221%
227%
245%
251%
258%
259%
262%
264%
271%
278%
284%
291%
295%
295%
327%
360%
365%
389%
391%
397%
402%
405%
407%
413%
415%
417%
420%
443%
452%
458%
458%
504%
512%
521%
560%
615%
637%
686%
686%
724%
808%
847%
869%
947%
1077%
Based on TPU review data: "Performance Summary" at 1920x1080
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

GPU Clock
1058 MHz
Boost Clock
1124 MHz
Memory Clock
900 MHz
1800 MHz effective

Memory

Memory Size
2 GB
Memory Type
DDR3
Memory Bus
128 bit
Bandwidth
28.80 GB/s

Render Config

Shading Units
384
TMUs
16
ROPs
16
SMM Count
3

Theoretical Performance

Pixel Rate
17.98 GPixel/s
Texture Rate
17.98 GTexel/s
FP32 (float) performance
863.2 GFLOPS
FP64 (double) performance
26.98 GFLOPS (1:32)

Board Design

Slot Width
Single-slot
Length
6.3 inches
160 mm
TDP
41 W
Outputs
1x DVI
1x DisplayPort
Power Connectors
None
Board Number
P2012

Graphics Features

DirectX
12.0 (11_0)
OpenGL
4.6
OpenCL
1.2
Vulkan
1.1.82
CUDA
5.0
Shader Model
5.1

GM107 GPU Notes

PureVideo HD: VP6
VDPAU: Feature Set E