Quadro 4000 and Tesla A100 - video card comparison

Quadro 4000

Tesla A100

Architecture

Fermi (2010−2014)

Ampere (2020−2022)

GPU code name

GF100

GA100

Market segment

Workstation

Workstation

Release date

November 2010

-

Interface

PCIe 2.0 x16

PCIe 4.0 x16

Supplementary power connectors

1x 6-pin

None

Core clock speed

475 MHz

1110 MHz

Floating-point performance

486.4 gflops

-

Manufacturing process technology

40 nm

7 nm

Number of transistors

3,100 million

38,000 million

Pipelines / CUDA cores

256

7936

Power consumption (TDP)

142 Watt

260 Watt

Texture fill rate

15.20

550.6

Memory type

GDDR5

HBM2e

Maximum RAM amount

2 GB

32 GB

Memory bus width

256 Bit

4096 Bit

Memory clock speed

2808 MHz

2400 MHz

Memory bandwidth

89.86 GB/s

1,229 GB/s

Display Connectors

1x DVI, 2x DisplayPort

No outputs

DirectX

12 (11_0)

12 (12_1)

CUDA

2.0

-

OpenCL

1.1

-

OpenGL

4.6

4.6

Shader Model

5.1

-

Vulkan

N/A

-