GeForce RTX 3090 and Radeon Pro W6900X - video card comparison

GeForce RTX 3090

Radeon Pro W6900X

Architecture

Ampere (2020−2022)

RDNA 2.0 (2020−2022)

GPU code name

Ampere GA102

Navi 21

Market segment

Desktop

Workstation

Release date

September 2020

August 2021

Interface

PCIe 4.0 x16

Supplementary power connectors

1x 12-pin

None

Boost clock speed

1700 MHz

2150 MHz

Core clock speed

1400 MHz

Manufacturing process technology

8 nm

7 nm

Number of transistors

28,300 million

26,800 million

Pipelines / CUDA cores

10496

5120

Power consumption (TDP)

350 Watt

300 Watt

Texture fill rate

556.0

688.0

Memory type

GDDR6X

GDDR6

Maximum RAM amount

24 GB

32 GB

Memory bus width

384 Bit

256 Bit

Memory clock speed

19500 MHz

16 GB/s

Memory bandwidth

936.2 GB/s

512.0 GB/s

Shared memory

Display Connectors

1x HDMI, 3x DisplayPort

1x HDMI, 4x Thunderbolt

HDMI

DirectX

12 Ultimate (12_2)

CUDA

8.5

OpenCL

2.0

2.1

OpenGL

4.6

Shader Model

6.5

Vulkan

1.2