GeForce RTX 3080 Ti Max-Q and GeForce RTX 4050 - video card comparison

GeForce RTX 3080 Ti Max-Q

GeForce RTX 4050

Architecture

Ampere (2020−2022)

Ada Lovelace

GPU code name

GA103S

AD107

Market segment

Desktop

Release date

January 2022

2023

Interface

PCIe 4.0 x16

PCIe 4.0 x8

Supplementary power connectors

None

1x 12-pin

Boost clock speed

1125 MHz

2640 MHz

Manufacturing process technology

8 nm

5 nm

Number of transistors

18,900 million

Pipelines / CUDA cores

7424

2560

Power consumption (TDP)

80 Watt

100 Watt

Texture fill rate

261.0

211.2

Memory type

GDDR6

Maximum RAM amount

16 GB

6 GB

Memory bus width

256 Bit

96 Bit

Memory clock speed

12 GB/s

18 GB/s

Memory bandwidth

384.0 GB/s

216.0 GB/s

Display Connectors

No outputs

1x HDMI 2.1, 3x DisplayPort 1.4a

HDMI

DirectX

12 Ultimate (12_2)

CUDA

8.6

8.9

OpenCL

3.0

OpenGL

4.6

Shader Model

6.5

6.7

Vulkan

1.3