Tesla K80 and GeForce RTX 3070 Ti GA102 - video card comparison

Tesla K80

GeForce RTX 3070 Ti GA102

Architecture

Kepler 2.0 (2013−2015)

Ampere (2020−2022)

GPU code name

GK210

GA102

Market segment

Workstation

Desktop

Release date

November 2014

October 2022

Interface

PCIe 3.0 x16

PCIe 4.0 x16

Supplementary power connectors

1x 8-pin

1x 12-pin

Boost clock speed

824 MHz

1770 MHz

Core clock speed

562 MHz

Manufacturing process technology

28 nm

8 nm

Number of transistors

7,100 million

28,300 million

Pipelines / CUDA cores

2496

6144

Power consumption (TDP)

300 Watt

290 Watt

Texture fill rate

171.4

339.8

Memory type

GDDR5

GDDR6X

Maximum RAM amount

12 GB

8 GB

Memory bus width

384 Bit

256 Bit

Memory clock speed

5012 MHz

19 GB/s

Memory bandwidth

240.6 GB/s

608.3 GB/s

Display Connectors

No outputs

1x HDMI 2.1, 3x DisplayPort 1.4a

HDMI

DirectX

12 (11_1)

12 Ultimate (12_2)

CUDA

3.7

8.6

OpenCL

1.2

3.0

OpenGL

4.6

Shader Model

5.1

6.6

Vulkan

1.1.126

1.3