IBM and GeForce RTX 3070 Max-Q - video card comparison

IBM

GeForce RTX 3070 Max-Q

Architecture

IBM (1999−2001)

Ampere (2020−2022)

GPU code name

IBM

GN20-E5 GA104

Market segment

Desktop

Laptop

Release date

January 2021

Interface

PCIe 4.0 x16

Supplementary power connectors

None

Boost clock speed

1290 MHz

Manufacturing process technology

250 nm

8 nm

Number of transistors

17,400 million

Pipelines / CUDA cores

5120

Power consumption (TDP)

80 Watt

Texture fill rate

237.4

Memory type

GDDR6

Maximum RAM amount

8 GB

Memory bus width

256 Bit

Memory clock speed

12 GB/s

Memory bandwidth

320.0 GB/s

Shared memory

Display Connectors

1x HDMI, 3x DisplayPort

HDMI

DirectX

6.0

12 Ultimate (12_2)

CUDA

8.6

OpenCL

N/A

2.0

OpenGL

1.2

4.6

Shader Model

6.5

Vulkan

N/A

1.2