GeForce RTX 2070 and Radeon Instinct MI250 - video card comparison

GeForce RTX 2070

Radeon Instinct MI250

Architecture

Turing (2018−2021)

CDNA 2.0 (2021)

GPU code name

Turing TU106

Aldebaran

Market segment

Desktop

Release date

August 2018

November 2021

Interface

PCIe 3.0 x16

PCIe 4.0 x16

Supplementary power connectors

1x 8-pin

2x 8-pin

Boost clock speed

1620 MHz

1700 MHz

Core clock speed

1410 MHz

Manufacturing process technology

12 nm

6 nm

Number of transistors

10,800 million

58,200 million

Pipelines / CUDA cores

2304

13312

Power consumption (TDP)

175 Watt

500 Watt

Texture fill rate

233.3

1,414

Memory type

GDDR6

HBM2e

Maximum RAM amount

8 GB

128 GB

Memory bus width

256 Bit

8192 Bit

Memory clock speed

14000 MHz

3.2 GB/s

Memory bandwidth

448.0 GB/s

3,277 GB/s

Shared memory

Display Connectors

1x DVI, 1x HDMI, 2x DisplayPort, 1x USB Type-C

No outputs

HDMI

G-SYNC support

DirectX

12 Ultimate (12_1)

N/A

CUDA

7.5

OpenCL

1.2

3.0

OpenGL

4.6

N/A

Shader Model

6.5

N/A

Vulkan

1.2.131

N/A

VR Ready