Tesla C2050 and Radeon Pro Duo Polaris - video card comparison

Tesla C2050

Radeon Pro Duo Polaris

Architecture

Fermi (2010−2014)

GCN 4.0 (2016−2020)

GPU code name

GF100

Ellesmere

Market segment

Workstation

Release date

July 2011

April 2017

Interface

PCIe 2.0 x16

PCIe 3.0 x16

Supplementary power connectors

1x 6-pin + 1x 8-pin

Core clock speed

574 MHz

1243 MHz

Floating-point performance

1,030.4 gflops

2x 5,728 gflops

Manufacturing process technology

40 nm

14 nm

Number of transistors

3,100 million

5,700 million

Pipelines / CUDA cores

448

2304

Power consumption (TDP)

238 Watt

250 Watt

Texture fill rate

32.14

179.0

Memory type

GDDR5

Maximum RAM amount

3 GB

16 GB

Memory bus width

384 Bit

256 Bit

Memory clock speed

3000 MHz

7000 MHz

Memory bandwidth

144.0 GB/s

224.0 GB/s

Display Connectors

1x DVI

1x HDMI, 3x DisplayPort

HDMI

DirectX

12 (11_0)

12 (12_0)

CUDA

2.0

OpenCL

1.1

2.0

OpenGL

4.6

Shader Model

5.1

6.4

Vulkan

N/A

1.2.131