Stream Processor and GeForce RTX 4070 Ti - video card comparison

Stream Processor

GeForce RTX 4070 Ti

Architecture

Ultra-Threaded SE

Ada Lovelace

GPU code name

R580

AD104-400

Market segment

Desktop

Release date

2006

January 2023

Interface

PCIe 1.0 x16

PCIe 4.0 x16

Supplementary power connectors

1x 6-pin

1x 12-pin

Boost clock speed

2610 MHz

Core clock speed

594 MHz

2310 MHz

Manufacturing process technology

90 nm

5 nm

Number of transistors

384 million

35,800 million

Pipelines / CUDA cores

7680

Power consumption (TDP)

165 Watt

285 Watt

Texture fill rate

9.504

626.4

Memory type

GDDR3

GDDR6X

Maximum RAM amount

1 GB

12 GB

Memory bus width

256 Bit

192 Bit

Memory clock speed

1296 MBps

21000 MHz

Memory bandwidth

41.47 GB/s

504.2 GB/s

Shared memory

Display Connectors

2x DVI, 1x S-Video

1x HDMI 2.1, 3x DisplayPort 1.4a

HDMI

DirectX

9.0c (9_3)

12 Ultimate (12_2)

CUDA

8.9

OpenCL

N/A

3.0

OpenGL

2.1 (full) Shader Model

3.06.7

Vulkan

N/A

1.3