GeForce RTX 3070 and Tesla P10 - video card comparison

GeForce RTX 3070

Tesla P10

Architecture

Ampere (2020−2022)

Pascal (2016−2021)

GPU code name

Ampere GA104

GP102

Market segment

Desktop

Workstation

Release date

September 2020

September 2016

Interface

PCIe 4.0 x16

PCIe 3.0 x16

Supplementary power connectors

1x 12-pin

1x 8-pin

Boost clock speed

1725 MHz

1493 MHz

Core clock speed

1500 MHz

-

Manufacturing process technology

8 nm

16 nm

Number of transistors

17,400 million

11,800 million

Pipelines / CUDA cores

5888

3840

Power consumption (TDP)

220 Watt

250 Watt

Texture fill rate

317.4

358.3

Memory type

GDDR6

GDDR5X

Maximum RAM amount

8 GB

24 GB

Memory bus width

256 Bit

384 Bit

Memory clock speed

14000 MHz

14.5 GB/s

Memory bandwidth

448.0 GB/s

694.3 GB/s

Shared memory

-

-

Display Connectors

1x HDMI, 3x DisplayPort

No outputs

HDMI

+

-

DirectX

12 Ultimate (12_2)

12 (12_1)

CUDA

8.5

6.1

OpenCL

2.0

1.2

OpenGL

4.6

4.6

Shader Model

6.5

6.4

Vulkan

1.2

1.2