GeForce RTX 3070 SUPER and GRID RTX T10-8 - video card comparison

GeForce RTX 3070 SUPER

GRID RTX T10-8

Architecture

Ampere (2020−2022)

Turing (2018−2021)

GPU code name

GA104

TU102

Market segment

Desktop

Workstation

Release date

Interface

PCIe 4.0 x16

PCIe 3.0 x16

Supplementary power connectors

1x 12-pin

1x 6-pin + 1x 8-pin

Boost clock speed

1695 MHz

1395 MHz

Manufacturing process technology

8 nm

12 nm

Number of transistors

18,000 million

18,600 million

Pipelines / CUDA cores

6144

4608

Power consumption (TDP)

250 Watt

260 Watt

Texture fill rate

325.4

401.8

Memory type

GDDR6

Maximum RAM amount

16 GB

8 GB

Memory bus width

256 Bit

384 Bit

Memory clock speed

16000 MHz

14 GB/s

Memory bandwidth

512.0 GB/s

672.0 GB/s

Display Connectors

1x HDMI, 3x DisplayPort

No outputs

HDMI

DirectX

12 Ultimate (12_2)

CUDA

7.5

OpenCL

3.0

OpenGL

4.6

Shader Model

6.7

Vulkan

1.3